Download - GAO-18-50, BORDER PATROL: Issues Related to Agent ...Issues Related to Agent Deployment Strategy and Immigration Checkpoints . What GAO Found . According to U.S. Border Patrol (Border

BORDER PATROL

Issues Related to Agent Deployment Strategy and Immigration Checkpoints

Report to Congressional Requesters

November 2017

GAO-18-50

United States Government Accountability Office

United States Government Accountability Office

Highlights of GAO-18-50, a report to congressional requesters

November 2017

BORDER PATROL Issues Related to Agent Deployment Strategy and Immigration Checkpoints

What GAO Found According to U.S. Border Patrol (Border Patrol), agent deployment decisions are based on factors such as staffing levels and the availability of agents, among other things. As of May 2017, nationwide, Border Patrol had about 1,900 fewer agents than authorized, which officials cited as a key challenge for optimal agent deployment. In recent years, attrition has exceeded hiring (an average of 904 agents compared to 523 agents) according to officials. GAO analyzed scheduling data, including time that agents were scheduled to be not working (for example, off duty or on leave) because these activities can affect deployment decisions by reducing the number of agents available on a particular day. GAO found that agents were available for deployment about 43 percent of the time.

Percentage of Agent Hours Scheduled for Time Off and Deployment Activities for the Southwest Border, Fiscal Years 2013–2016

From fiscal years 2012 through 2016, Border Patrol apprehended a total of almost 2 million individuals along the southwest border, and these apprehensions increasingly occurred closer to the border, with 42 percent of apprehensions occurring one-half mile or less from the border in fiscal year 2016 compared to 24 percent in fiscal year 2012. One driver for this change is the increasing number of apprehensions of children, whom officials report may turn themselves in to Border Patrol without attempting to evade detection. Meanwhile, over this period, the locations where seizures of contraband occurred remained roughly the same, with the majority occurring 10 or more miles from the border.

For fiscal years 2013 through 2016, GAO found that 2 percent of apprehensions and 43 percent of seizures occurred at checkpoints; however, determining the extent to which apprehensions and seizures are attributable to checkpoints is difficult because of long-standing data issues. More apprehensions and seizures may be attributable to checkpoints, but Border Patrol’s reporting does not distinguish apprehensions that occurred “at” versus “around” a checkpoint. Border Patrol is drafting guidance to clarify how checkpoint apprehension and seizure data are to be recorded that would respond to a 2009 GAO recommendation to improve the internal controls for management oversight of checkpoint data. GAO also determined that seizures at checkpoints differed from those at other locations. Specifically, 40 percent of seizures at checkpoints were 1 ounce or less of marijuana from U.S. citizens. In contrast, seizures at other locations were more often higher quantities of marijuana seized from aliens.

View GAO-18-50. For more information, contact Rebecca Gambler at (202) 512-8777 or [email protected].

Why GAO Did This Study The Border Patrol has primary responsibility for securing the border between U.S. ports of entry. On the southwest border, Border Patrol deploys agents along the immediate border and in areas up to 100 miles from the border as part of a layered approach known as the defense in depth strategy. Immigration checkpoints, generally located between 25 and 100 miles from the border, are one element of this strategy. GAO was asked to review the defense in depth strategy.

This report addresses: (1) the factors Border Patrol considers in deploying agents, (2) where apprehensions of illegal crossers and seizures of contraband are occurring, and (3) what data show about how checkpoints contribute to apprehensions and seizures, among other objectives. To answer these questions, GAO analyzed Border Patrol documents and data on apprehensions and seizures from fiscal year 2012 through 2016, visited two southwest border sectors, interviewed officials from the other seven southwest border sectors and Border Patrol headquarters, and reviewed prior GAO work on border security.

What GAO Recommends GAO is not making any new recommendations at this time but has previously recommended that Border Patrol establish internal controls for checkpoint data, among other things. DHS concurred with this recommendation and has taken some steps to improve the quality of checkpoint data, but additional actions are needed to fully implement the recommendation.

http://www.gao.gov/products/GAO-18-50


mailto:[email protected]

Page i GAO-18-50 Border Patrol

Letter 1

Background 6 Border Patrol Deploys Agents Based on Availability and

Geography, Among Other Factors, and Agent Activity Schedules Vary By Sector 16

Apprehensions Occurred Closer to the Border in Fiscal Year 2016 Compared to Fiscal Year 2012, While Seizure Locations Remained Relatively Unchanged 30

Long-Standing Data Quality Issues Make It Difficult to Precisely Measure Checkpoints’ Contributions to Apprehensions and Seizures 36

Collecting Additional Data Could Improve Border Patrol’s Existing Efforts to Identify and Respond to Community Concerns Regarding the Defense in Depth Strategy 49

Agency Comments 56

Appendix I Status of GAO Recommendations Related to Checkpoints 57

Appendix II Trends in Southwest Border Apprehensions, Fiscal Years 2012 through 2016 63

Appendix III Trends in Southwest Border Seizures, Fiscal Years 2012 through 2016 72

Appendix IV Comments from the Department of Homeland Security 82

Appendix V GAO Contact and Staff Acknowledgments 86

Tables

Table 1: Number of Authorized and On-Board Agents, by Southwest Border Sector, as of May 2017 9

Table 2: Agent Hours Scheduled for Southwest Border Checkpoints, by Sector, Fiscal Years 2013–2016 29

Contents

Page ii GAO-18-50 Border Patrol

Table 3: Southwest Border and Rio Grande Valley Sector Apprehensions from Mexico and Select Central American Countries, Fiscal Years (FY) 2012 and 2016 33

Table 4: Categories Describing the Relationship between Apprehension and Seizure (Event) Locations and Checkpoints 39

Table 5: Southwest Border Checkpoint Apprehensions and Seizures by Location Category, Fiscal Years 2013–2016 41

Table 6: Southwest Border Apprehensions by Sector and by Distance from the Border for Fiscal Years (FY) 2012–2016 65

Table 7: Southwest Border Checkpoint Apprehensions by Sector and by Location Category, Fiscal Years 2013–2016 68

Table 8: Type and Number of Southwest Border Seizures, Fiscal Years (FY) 2012–2016 72

Table 9: Southwest Border Seizures by Sector and by Distance from the Border for Fiscal Years (FY) 2012–2016 75

Table 10: Southwest Border Seizures by Sector and by Location Category, Fiscal Years 2013–2016 77

Table 11: Southwest Border Marijuana Seizures by Quantity Seized at Checkpoints and at Non-Checkpoint Locations, Fiscal Years 2013–2016 81

Figures

Figure 1: Border Patrol Sectors along the Southwest Border and Border Stations and Zones in the Tucson Sector 7

Figure 2: Images of a Forward Operating Base (FOB) and Checkpoints 11

Figure 3: Permanent Checkpoints near the Southwest Border 13 Figure 4: Checkpoint Canine Team Inspects a Vehicle 14 Figure 5: Percentage of Agent Hours Scheduled for Time Off and

Deployment Activities for the Southwest Border, Fiscal Years 2013–2016 19

Figure 6: Distribution of Border Patrol Agents’ Scheduled Work Time, by Sector, Fiscal Years 2013–2016 26

Figure 7: Percentage of Border Patrol Agents’ Time for Operations and Patrol Scheduled for Different Zone Types, Fiscal Years 2013–2016 28

Figure 8: Distribution of Southwest Border Apprehensions by Distance from the Border, Fiscal Years 2012–2016 31

Page iii GAO-18-50 Border Patrol

Figure 9: Distribution of Southwest Border Seizures by Distance from the Border, Fiscal Years 2012–2016 34

Figure 10: Percentage of Seizures by Distance From the Border for the Tucson, Big Bend, and Rio Grande Valley Sectors, Fiscal Years 2012–2016 35

Figure 11: Maps of Apprehensions Surrounding Two Checkpoints, Fiscal Years 2013–2016 42

Figure 12: Breakdown of Seizures by Type of Seizure, Location of Seizure, and Individuals Associated with the Seizure, Fiscal Years 2013–2016 47

Figure 13: Mangled Gate Suspected to Have Been Caused by Illegal Border Crossers along Highway 281 in Brooks County, Texas 51

Figure 14: Southwest Border Apprehensions by Border Patrol Sector, Fiscal Years 2012–2016 64

Figure 15: Southwest Border Seizures by Border Patrol Sector, Fiscal Years 2012–2016 74

Abbreviations BCL Border Community Liaison Border Patrol U.S. Border Patrol BPAPRA Border Patrol Agent Pay Reform Act of 2014 BPETS Border Patrol Enforcement Tracking System CBP U.S. Customs and Border Protection CCMS Compliments and Complaints Management System checkpoint immigration checkpoint CPMO Checkpoint Program Management Office DHS Department of Homeland Security FOB Forward Operating Base FY fiscal year GPS Global Positioning System

This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately.

GAO-18-50 Border Patrol

441 G St. N.W. Washington, DC 20548

November 8, 2017

The Honorable Claire McCaskill Ranking Member Committee on Homeland Security and Governmental Affairs U.S. Senate

The Honorable Martha McSally Chairwoman Subcommittee on Border and Maritime Security Committee on Homeland Security House of Representatives

The United States’ international border with Mexico continues to be vulnerable to illegal cross-border traffic. The U.S. Border Patrol (Border Patrol)—an office within the Department of Homeland Security’s (DHS) U.S. Customs and Border Protection (CBP) that has primary responsibility for securing the border between the U.S. ports of entry1—recorded more than 400,000 apprehensions and more than 15,000 seizures in fiscal year 2016.2 Agents are among the key resources Border Patrol deploys—along with infrastructure and technology—to respond to cross-border threats, and Border Patrol’s most recent strategic plan calls for a risk-based approach to resource deployment.

1Ports of entry are facilities that provide for the controlled entry into or departure from the United States. Specifically, a port of entry is any officially designated location (seaport, airport, or land border location) where DHS officers or employees are assigned to clear passengers and merchandise, collect duties, and enforce customs laws, and where DHS officers inspect persons entering or applying for admission into, or departing the United States pursuant to U.S. immigration law. 2See 6 U.S.C. § 211(a) (establishing CBP within DHS), (c) (enumerating CBP’s duties), (e) (establishing and listing duties of U.S. Border Patrol within CBP). Apprehensions refer to arrests of aliens (persons who are not U.S. citizens or nationals; i.e., foreign nationals) who are potentially removable on grounds of inadmissibility or deportability. We refer to these individuals in this report as potentially removable aliens. Border Patrol agents may also encounter and, as appropriate, arrest U.S. citizens or non-removable aliens for violating U.S. law, such as smuggling contraband. In this report, a seizure refers to the confiscation of one type of contraband from an individual (who may be a foreign national or a U.S. citizen and who is arrested in conjunction with the seizure). There may be multiple seizures associated with a single arrest. For example, Border Patrol could seize both marijuana and cocaine from the same individual, and for the purposes of this report that would be considered two seizure events.

Letter


Along the southwest border, Border Patrol deploys agents along the immediate border and in areas up to 100 miles from the border as part of a layered approach the agency refers to as the defense in depth strategy. As part of this strategy, Border Patrol deploys agents on temporary duty assignments to Forward Operating Bases (FOBs) located in difficult to access areas near the border to reduce the response time to threats in these areas, and it also deploys agents to immigration checkpoints (checkpoints) that are generally located on highways 25 to 100 miles from the border.

At checkpoints, Border Patrol agents screen vehicles for aliens who were able to illegally cross the border undetected at or between ports of entry. Border Patrol checkpoints are located on major U.S. highways and secondary roads. This permits checkpoints to be far enough inland to detect and apprehend aliens in violation of U.S. immigration law, smugglers, and potential terrorists attempting to travel farther into the interior of the United States on ingress routes after evading detection or otherwise avoiding required inspection at the border. According to Border Patrol data, there are about 140 locations that Border Patrol uses for checkpoint operations near the southwest border, although not all of them are operational at all times.3 Some of these checkpoints have a permanent structure with off-highway inspection lanes and technology to facilitate inspection and convenience, while other checkpoints have temporary infrastructure in the form of trailers and generators. DHS’s Congressional Justification for the fiscal year 2018 budget notes that, because checkpoints are an integral part of the defense in depth strategy, measurements of checkpoint events (apprehensions and seizures) serve as barometers of the effectiveness of Border Patrol’s overall strategy to deny illegal entries into the United States.

Communities located near the southwest border may be positively or negatively impacted by the placement and operation of checkpoints and other Border Patrol resources deployed under the defense in depth strategy. You requested that we review Border Patrol’s defense in depth strategy. This report addresses the following questions:

1. What factors does Border Patrol consider in deploying agents to the different layers of the defense in depth strategy along the southwest border, including FOBs along the border and checkpoints?

3Border Patrol also has checkpoints near the northern border.


2. Where are apprehensions and seizures occurring in relation to the border, as a reflection of Border Patrol’s defense in depth strategy?

3. What do data show about how checkpoints contribute to apprehensions and seizures?

4. What is known about the effects of Border Patrol’s defense in depth strategy on surrounding communities, and how, if at all, does Border Patrol engage with these communities to identify and respond to any such effects?

To address these objectives, we analyzed Border Patrol data and documents, visited two southwest border sectors, and interviewed Border Patrol officials from the other seven southwest border sectors and Border Patrol headquarters.4 Specifically, we analyzed Border Patrol data on planned agent deployment, apprehensions, and seizures across all nine southwest border sectors from fiscal years 2012 through 2016. We selected these years because they were the five most recent fiscal years for which complete data were available at the time of our review.

We also selected the two sectors to visit (Tucson sector and Rio Grande Valley sector) based on the role of checkpoints in those locations, the presence of FOBs, the role of known community groups that engage with Border Patrol, and a relatively high level of known cross-border illegal activity in those sectors, among other factors. In these two sectors, we met with sector and station officials, local law enforcement agencies, and community groups. We selected local law enforcement agencies and community groups based on recommendations from sector officials and geographic considerations, such as proximity to checkpoints.5 We also visited Border Patrol facilities including checkpoints and FOBs. Both in-person during our visits to the two selected sectors and by phone with the remaining seven sectors, we interviewed officials responsible for making agent deployment decisions to discuss how sectors and stations decide to deploy agents, the operational results of agent deployment, and how

4Border Patrol divides responsibility for southwest border security operations geographically among nine sectors, each with its own sector headquarters. Each sector is further divided into varying numbers of stations each with its own area of responsibility. 5In the Tucson sector, we met with the Arizona Department of Public Safety, Cochise County Sheriff’s Office, Pima County Sheriff’s Department, Sahuarita Police Department, Santa Cruz County Sheriff’s Office, Douglas Unified School District, and the Santa Cruz Valley Citizens Council. In the Rio Grande Valley sector, we met with the Brooks County Sheriff’s Office, Hidalgo County Sheriff’s Office, and the South Texans’ Property Rights Association.


sectors and stations respond to any effects of Border Patrol’s defense in depth strategy on the surrounding community. We also interviewed officials from Border Patrol headquarters—specifically officials from the Checkpoint Program Management Office (CPMO) and from the Border Community Liaison (BCL) Program—and officials responsible for overseeing and analyzing Border Patrol deployment, apprehension, and seizure data. These interviews focused on Border Patrol’s oversight of and data regarding agent deployment, enforcement actions, and checkpoints.

We also reviewed our prior work regarding checkpoints, the defense in depth strategy, and issues related to private property damage along the southwest border.6 In particular, we reviewed recommendations from our August 2009 report on checkpoints and assessed actions taken in response to these recommendations to determine whether they had been implemented.

To address our first objective, we analyzed data on planned agent deployment from the Border Patrol Enforcement Tracking System (BPETS) for fiscal years 2013 through 2016.7 We reviewed the activities in the data to determine broad categories represented among the activities.8 Border Patrol officials responsible for overseeing the data concurred with our categories and how we organized the activities into these categories. We assessed the reliability of these data by testing for missing data and obvious errors, reviewing related documentation, and 6GAO, Border Patrol: Checkpoints Contribute to Border Patrol’s Mission, but More Consistent Data Collection and Performance Measurement Could Improve Effectiveness, GAO-09-824 (Washington, D.C.: Aug. 31, 2009); Border Patrol: Key Elements of New Strategic Plan Not Yet in Place to Inform Border Security Status and Resource Needs, GAO-13-25 (Washington, D.C.: Dec. 10, 2012); and Southwest Border: Issues Related to Private Property Damage, GAO-15-399 (Washington, D.C.: Apr. 30, 2015). 7We requested and received planned scheduling data for fiscal year 2012, but fiscal year 2012 data were not comparable to subsequent years’ data because Border Patrol modified its approach to scheduling in March 2012. In particular, prior to this time, scheduling activities were recorded in 1-day increments, with the ability to assign one activity per day. Data recorded after the change were recorded in hours and an agent’s work shift could be split into multiple activities with hours assigned to each activity. 8To determine categories for this analysis, we reviewed BPETS data to identify themes among the activities in the data. We proposed categories and the assignment of activities to these categories to Border Patrol officials for the purpose of our analysis. We sought input from and reached agreement with officials who oversee the BPETS data and the Personal Requirements Determination project regarding the categories and how activities were organized into the categories.





interviewing knowledgeable agency officials. We determined that the data used in our analyses were sufficiently reliable for the purposes of this report.

To address our second and third objectives, we obtained and analyzed apprehension and seizure data from Border Patrol’s Enforcement Integrated Database/e3 (e3) database for fiscal years 2012 through 2016. We assessed the reliability of these data by testing for missing data and obvious errors, reviewing related documentation, and interviewing knowledgeable agency officials. We used Global Positioning System (GPS) coordinates contained in the e3 data to analyze distance from the border and proximity and relation to checkpoints for apprehensions and seizures. In assessing the reliability of these data, we identified some apprehension and seizure records with GPS coordinates that were outside the scope of our review. We excluded the small percentage of records that had missing GPS coordinates or had GPS coordinates that were not within the boundaries of southwest Border Patrol sectors.9 We determined that the data used in our analyses were sufficiently reliable for the purposes of this report.10 For our analysis of apprehensions’ and seizures’ proximity and relation to checkpoints, we analyzed data for

9Some GPS coordinates were recorded as being outside of the United States. GPS coordinates for the U.S. southwest border region have a positive number for latitude (northern hemisphere) and a negative number for longitude (western hemisphere). If the latitude coordinate was recorded with a negative sign (indicating the southern hemisphere), we removed the negative sign. Similarly, if the longitude coordinate excluded the negative sign (indicating the eastern hemisphere), we added the negative sign. For any such adjustments we made, we determined if the adjusted coordinates would be within the United States and a southwest border sector. We included records with adjusted coordinates that were within southwest border sectors, and excluded the records that remained outside the United States and/or southwest border sectors. After making such adjustments, we identified 11,710 apprehensions and 443 seizures that had missing GPS coordinates, had coordinates outside the United States, and/or coordinates outside the boundaries of southwest border sectors, and we therefore excluded these data from our analyses. 10The GPS coordinates were sufficiently reliable for determining the locations of apprehensions and seizures. Using the GPS coordinates, we mapped the locations of apprehensions, seizures, and checkpoints. We analyzed these maps to assess the reliability of other data fields that Border Patrol uses to attribute apprehensions and seizures to checkpoints. The results of that analysis revealed data quality issues for the other data fields related to checkpoints, and the issues related to these other data fields are discussed in detail later in this report.


fiscal years 2013 through 2016 because a listing of checkpoints that were operational in fiscal year 2012 was not available.11

To address our fourth objective, we reviewed BCL program documents from headquarters and all nine southwest border sectors, interviewed officials in these sectors, and met with community members and local law enforcement officials in the two sectors we visited. We spoke with officials from all nine southwest border sectors about their views on the effects that Border Patrol and illegal cross-border activity have had on nearby communities up to 100 miles from the border. Our interviews also focused on Border Patrol’s efforts to identify and respond to any effects, including the organization of and actions taken as part of the BCL program. Although the information we obtained from these interviews cannot be generalized to other communities near border checkpoints, these interviews provided important insights and perspectives about the checkpoints.

We conducted this performance audit from July 2016 to November 2017 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives.

Border Patrol divides responsibility for southwest border security operations geographically among nine sectors, each with its own sector headquarters. Each sector is further divided into varying numbers of stations. For example, the Tucson sector has divided geographic responsibility across eight stations, seven of which have responsibility for miles of land directly on the U.S.-Mexico border. Stations’ areas of responsibility are divided into zones. Border Patrol refers to “border zones”—those having international border miles—and “interior zones”—

11At the end of each fiscal year, Border Patrol creates a record of checkpoints that were potentially operational during that year. Border Patrol began documenting these lists of checkpoints in fiscal year 2013, and therefore no such list for fiscal year 2012 was available for the purposes of our analysis.

Background

Border Patrol’s Organizational Structure


those without international border miles. For example, as shown in figure 1, within the Tucson sector, the Sonoita station has only border zones, the Willcox station has only interior zones, and the other stations have a mix of both border and interior zones. According to Border Patrol officials, dividing stations into zones allows sectors to more effectively analyze border conditions, including terrain, when planning how to deploy agents. Zone dimensions are largely determined by geography and topographical features, and zone size can vary significantly.

Figure 1: Border Patrol Sectors along the Southwest Border and Border Stations and Zones in the Tucson Sector


In fiscal years 2011 through 2016, Border Patrol was statutorily required to maintain a minimum of 21,370 full-time equivalent agent positions, but Border Patrol has faced challenges in staffing to that minimum level.12 As of May 2017, Border Patrol had about 19,500 agents on board. Additionally, in January 2017, an executive order called for the hiring of 5,000 additional Border Patrol agents, subject to available appropriations, and Border Patrol is aiming to have 26,370 agents by the end of fiscal year 2021.13 The Acting Commissioner of CBP reported in a February 2017 memo to the Deputy Secretary for Homeland Security that from fiscal year 2013 to fiscal year 2016, Border Patrol hired an average of 523 agents per year while experiencing a loss of an average of 904 agents per year.14 The memo cited challenges such as competing with other federal, state, and local law enforcement organizations for applicants. In particular, the memo noted that CBP faces hiring and retention challenges compared to DHS’s U.S. Immigration and Customs Enforcement (which is also planning to hire additional law enforcement personnel) because CBP’s hiring process requires applicants to take a polygraph examination, Border Patrol agents are deployed to less desirable duty locations, and Border Patrol agents generally receive lower compensation.

12Department of Defense and Full-Year Continuing Appropriations Act, 2011, Pub. L. No. 112-10, div. B, tit. VI, § 1608, 125 Stat. 38, 140; Consolidated Appropriations Act, 2012, Pub. L. No. 112-74, div. D, tit. II, 125 Stat. 786, 946 (2011); Consolidated and Further Continuing Appropriations Act, 2013, Pub. L. No. 113-6, div. D, tit. II, 127 Stat. 198, 345; Consolidated Appropriations Act, 2014, Pub. L. No. 113-76, div. F, tit. II, 128 Stat. 5, 249; Department of Homeland Security Appropriations Act, 2015, Pub. L. No. 114-4, tit. II, 129 Stat. 39, 41; Consolidated Appropriations Act, 2016, Pub. L. No. 114-113, div. F, tit. II, 129 Stat. 2242, 2495 (2015). For fiscal year 2017, Border Patrol was not mandated to have a minimum number of agents, but Congress directed CBP to continue working to develop a fully justified workforce staffing model that would provide validated requirements for all U.S. borders and to brief the appropriations committees on its progress in this regard within 30 days of the enactment of the Consolidated Appropriations Act, 2017. See Explanatory Statement (163 Cong. Rec. H3327, H3809-10 (daily ed. May 3, 2017)) accompanying Pub. L. No. 115-31, 131 Stat. 135. 13Border Security and Immigration Enforcement Improvements, Exec. Order No. 13767, § 8, 82 Fed. Reg. 8793, 8795 (Jan. 30, 2017). 14The Acting Commissioner’s memo outlines plans and requests to assist Border Patrol in hiring more agents, including the additional 5,000 agents called for in the Executive Order on Border Security and Immigration Enforcement Improvements. We have ongoing work regarding hiring and retention challenges at CBP. This work is being conducted for the ranking members of the House Committee on Homeland Security and the Subcommittees on Border and Maritime Security and on Oversight and Management Efficiency.

Staffing Levels and Agent Deployment


Border Patrol headquarters, with input from the sectors, determines how many authorized agent positions will be allocated to each of the sectors. According to Border Patrol officials, these decisions take into account the relative needs of the sectors, based on threats, intelligence, and the flow of illegal activity. Each sector’s leadership determines how many of the authorized agent positions will be allocated to each station within their sector. Sector leadership also distributes newly assigned agents—those agents recently hired whom headquarters has assigned to the sector, or existing agents who are being transferred—to specific stations within the sector. Table 1 shows the number of authorized agent positions for each southwest border sector as well as the number of agents who were assigned to each of those sectors, as of May 2017.

Table 1: Number of Authorized and On-Board Agents, by Southwest Border Sector, as of May 2017

Sector Number of authorized agent positions

Number of on-board agents

Difference

San Diego 2,484 2,262 222 El Centro 1,120 888 232 Yuma 792 843 -51 Tucson 3,825 3,716 109 El Paso 2,415 2,193 222 Big Bend 640 495 145 Del Rio 1,642 1,398 244 Laredo 1,852 1,584 268 Rio Grande Valley 3,201 3,143 58 Total 17,971 16,522 1,449

Source: Border Patrol data. | GAO-18-50

Once a sector assigns agents to a station, station officials assign agents to a shift. Most agents work 10-hour shifts, which allows for some overlap


in time for the outgoing shift to relay key information to the incoming shift.15 Most agents work 5 days per week with 2 off duty days.

Border Patrol has 17 FOBs that are established in forward or remote locations in five of the nine southwest border sectors to sustain Border Patrol operations.16 According to Border Patrol officials, the primary function of these facilities is to give the Border Patrol a tactical advantage by reducing response time to threats or actionable intelligence. Typically, agents are assigned for 7 days, during which they reside at the FOB and deploy to their assigned duties.17 FOBs allow agents to be pre-positioned at these locations, which reduces the portion of an agent’s shift that is spent in transit between the station and the patrol location. In addition, these facilities are intended to increase security awareness and presence in the border areas where they are located. FOBs are staffed by Border Patrol agents on temporary duty assignments from their permanent duty station. After their shift, they are normally required to remain at the FOB to rest, prepare for their next shift, and be available, if needed, to respond to operational issues. Figure 2 includes a photo of a FOB in the Rio Grande Valley sector. Five of the nine southwest border sectors—Yuma, Tucson, El Paso, Big Bend, and the Rio Grande Valley—have FOBs, whereas the other four sectors—San Diego, El Centro, Del Rio, and Laredo—do not.

15The Border Patrol Agent Pay Reform Act of 2014 (BPAPRA) established the current overtime compensation system for Border Patrol agents, in which agents individually elect and are subsequently assigned by the agency to one of three rates of pay commensurate with the amount of scheduled overtime the agents elect or are assigned to work. See Pub. L. No. 113-277, 128 Stat. 2995 (2014) (amending titles 5 and 29 of the United States Code with respect to overtime compensation for Border Patrol agents). In general, BPAPRA provides that at least 90 percent of all Border Patrol agents at each duty location shall be assigned to the rate of pay that corresponds with a 10-hour daily schedule for 5 workdays per week, yielding a 25 percent overtime supplement. 16Some of these facilities are permanent structures while others, also known as camps, are more temporary in nature. Border Patrol has 12 permanent FOBs and 5 camps, and we refer to these facilities collectively as FOBs in this report. 17DHS Office of the Inspector General, Conditions at CBP’s Forward Operating Bases along the Southwest Border, OIG-16-37 (Washington, D.C.: Feb. 8, 2016).

FOBs and Checkpoints


Figure 2: Images of a Forward Operating Base (FOB) and Checkpoints


Border Patrol operates two types of checkpoints—permanent and tactical—that differ in terms of size, infrastructure, and location. While both types of checkpoints are generally operated at fixed locations, permanent checkpoints—as their name suggests—are characterized by their brick and mortar structures, that may include off-highway covered lanes for vehicle inspection and several buildings including those for administration, detention of persons suspected of smuggling or other illegal activity, and kennels for canines used in the inspection process.18 Figure 2 shows examples of permanent and tactical checkpoints we observed in the Rio Grande Valley and Tucson sectors, and figure 3 is a map depicting the locations of permanent checkpoints near the southwest border.

18The Tucson sector is the only sector along the southwest border without permanent checkpoints. Although other sectors along the southwest border deploy a combination of permanent and tactical checkpoints, the Tucson sector has only tactical checkpoints that operate from fixed locations. Legislation effectively prohibited the construction of permanent checkpoints in the Tucson sector, beginning in fiscal year 1999. Specifically, the Omnibus Consolidated and Emergency Supplemental Appropriations Act, 1999, stated that “no funds shall be available for the site acquisition, design, or construction of any Border Patrol checkpoint in the Tucson sector.” See Pub. L. No. 105-277, 112 Stat. 2681, 2681-59 (1998).The effect of this legislative language was that no permanent checkpoints could be planned or constructed in this sector, which had no permanent checkpoints when the prohibition took effect.


Figure 3: Permanent Checkpoints near the Southwest Border

Border Patrol agents at checkpoints have legal authority that agents do not have when conducting roving patrols away from the border. The United States Supreme Court ruled that Border Patrol agents may stop a vehicle at fixed checkpoints for brief questioning of its occupants even if there is no reason to believe that the particular vehicle contains illegal entrants, and also held that the operation of a fixed checkpoint does not require a judicial warrant.19 The Court further held that, provided the intrusion is sufficiently minimal so as not to require particularized justification, Border Patrol agents “have wide discretion” to refer motorists selectively to a secondary inspection area for additional brief questioning.20 In contrast, the Supreme Court held that Border Patrol agents on roving patrol may stop a vehicle only if they have reasonable suspicion that the vehicle contains aliens who may be illegally in the 19U.S. v. Martinez-Fuerte, 428 U.S. 543, 545 (1976). 20Id. at 563-4.


United States—a higher threshold for stopping and questioning motorists than at checkpoints.21 The constitutional threshold for searching a vehicle is the same, however, and must be supported by either consent or probable cause, whether in the context of a roving patrol or a checkpoint search.22 Probable cause can include a canine detecting something it is trained to detect (e.g., concealed people, narcotics). Figure 4 shows a Border Patrol canine team inspecting a vehicle at a checkpoint.

Figure 4: Checkpoint Canine Team Inspects a Vehicle

We have previously reported on topics related to the defense in depth strategy, and specifically on checkpoints.23 In August 2009, we reported on the measurement of checkpoint performance and the impact of checkpoint operations on nearby communities, among other things

21U.S. v. Brignoni-Ponce, 422 U.S. 873, 881-2 (1975). 22U.S. v. Ortiz, 422 U.S. 891, 896-8 (1975). 23See GAO-13-25, GAO-09-824, and GAO, Border Patrol: Available Data on Interior Checkpoints Suggest Differences in Sector Performance, GAO-05-435 (Washington, D.C.: July 22, 2005).

Previous GAO Work





related to checkpoints.24 In that report, we made recommendations to, among other things, strengthen checkpoint design and staffing and improve the measurement and reporting of checkpoint effectiveness, including measuring community impacts. CBP has implemented two of our recommendations from that report—specifically, Border Patrol explored and considered the feasibility of a checkpoint performance model and required that traffic volumes be studied and considered when designing new permanent checkpoints. Appendix I provides details on the status of all six recommendations from that report. We also reported in December 2012 on how Border Patrol manages personnel resources at the southwest border, including aspects of the defense in depth strategy, such as where apprehensions and seizures were occurring relative to the southwest border. That report focused on the Tucson sector—which at the time had the most Border Patrol apprehensions of the nine southwest border sectors—and compared data on agent deployment, apprehensions, and seizures from the Tucson sectors with data for other sectors.25

24GAO-09-824. 25In that report, we also reported on Border Patrol’s implementation of its 2012–2016 strategic plan. We made recommendations that Border Patrol develop milestones and time frames for developing border security goals and measures to assess progress made and resource needs to support the implementation of the strategic plan. Border Patrol has not yet fully implemented these recommendations. See GAO-13-25.




According to sector officials, decisions about agent deployment in terms of location and activity are based on multiple factors, including the availability of agents for a given shift, the geography in a station’s area of responsibility, and illegal traffic patterns. For example, when considering the various assignments that need to be filled for a given day, supervisors must take into account agents that are unavailable because they are off duty, on scheduled leave, or are scheduled to attend training. The geography in a station’s area of responsibility can also affect decisions about where to deploy available agents. For example, Border Patrol may have limited access to certain areas because of challenging terrain, limited or poor quality roads, or private ownership. Supervisors also review information about illegal traffic patterns in their areas of responsibility to determine where enforcement operations may be needed.

One key factor in how Border Patrol makes deployment decisions at the station level is the overall number of agents available. Officials from all nine southwest border sectors cited current staffing levels and the availability of agents as a challenge for optimal deployment. Nationwide, as of May 2017, Border Patrol had nearly 1,900 fewer agents than authorized and has faced hiring and retention challenges in recent years. As shown in table 1 earlier, eight of the nine southwest border sectors were below their authorized agent staffing levels as of May 2017.26 As

26As of May 2017, Yuma sector had 51 agents more than its authorized amount. Yuma sector officials said that this does not result in being overstaffed because the workload for the sector has increased as a result of increased illegal traffic, and thus the sector has sought additional authorized positions to account for the heightened risk in the sector.

Border Patrol Deploys Agents Based on Availability and Geography, Among Other Factors, and Agent Activity Schedules Vary By Sector Border Patrol Assigns Agents Based on Factors Such As Availability, Geography, and Illegal Traffic Patterns

Number and Availability of Agents


such, resources are constrained and station officials must make decisions about how to prioritize activities for deployment given the number of agents available.

Within sectors, some stations may be comparatively more understaffed than others because of recruitment and retention challenges, according to officials. Generally, sector officials said that the recruitment and retention challenges associated with particular stations were related to quality of life factors in the area near the station—for example, agents may not want to live with their families in an area without a hospital, with low-performing schools, or with relatively long commutes from their homes to their duty station. This can affect retention of existing agents, but it may also affect whether a new agent accepts a position in that location. For example, officials in one sector said that new agent assignments are not based solely on agency need, but rather also take into consideration agent preferences. These officials added that there is the potential that new agents may decline offers for stations that are perceived as undesirable, or they may resign their position earlier than they otherwise would to pursue employment in a more desirable location.

Supervisors make decisions about how to deploy agents based on the number of agents assigned to a shift who are available to work the shift on a particular day. On any given day, some agents will be off duty, in training, or have annual or sick leave scheduled, thereby reducing the number of agents available for deployment during a shift.

To assess how Border Patrol has scheduled and deployed agents across the southwest border sectors, we analyzed the scheduled deployment data that supervisors entered into BPETS for fiscal years 2013 through 2016. Supervisors enter data into BPETS in advance of a shift to track expected time and attendance. Supervisors record work status by indicating whether an agent will be working, off duty, or otherwise not working (for example, on annual leave or scheduled sick leave), and for agents who are working, the supervisors also record an assignment to which the agent is expected to be deployed that day. We analyzed these assignments to determine how agents’ work time was distributed among activities in the following categories:27

27Supervisors record scheduled hours in BPETS based on a combination of four data fields: (1) work status, (2) activity, (3) subactivity, and (4) assignment. In coordination with Border Patrol officials, we paired the activity and subactivity options in BPETS and assigned each pair to the higher-order categories.


• Operations and Patrol refers to frontline activities that involve identifying and apprehending illegal entrants and identifying and seizing contraband. Some specific examples include linewatch, sign cutting, and checkpoint duties.28

• Operational Support refers to activities, such as intelligence gathering or surveillance, that support frontline agents conducting operations and patrol activities.

• Processing refers to activities that occur after apprehending an individual, including transport, processing, detention, and removal.

• Legal Support and Litigation refers to activities, such as attending court proceedings, that involve prosecution of apprehended individuals.

• Training refers to activities that involve providing instruction, attending training, or completing qualification/certification tests.

• Administrative and Other Non-Enforcement Activities refers to activities other than those above, including public relations, hiring and recruitment, and policy and compliance.

Our analysis included time that agents were scheduled to be off duty or on scheduled leave because scheduled time off can affect supervisors’ deployment decisions by reducing the number of agents available on a particular day.

As shown in figure 5, agents were unavailable for deployment for a total of 42 percent of time (off duty time, scheduled non-work time, and training), and about 43 percent of agents’ time was scheduled for operations and patrol activities in the field.29 As an example, this means that a station with 300 total agents—with 100 agents assigned to each of three shifts—would have had on average about 42 of the 100 agents per shift unavailable because of planned time not working (off duty or other scheduled non-work time) or in training. Of the remaining 58 agents, on average, about 43 would have been scheduled to field-based operations and patrol activities, and 15 would have been assigned to other activities. 28Linewatch involves maintaining surveillance from a covert position to detect, prevent, or apprehend illegal entrants at or near the land border. Sign cutting involves following tracks, marks, and other physical evidence to detect illegal entrants. 29Border Patrol provided us with scheduling data for fiscal year 2012; however the data are not comparable to more recent years because in March 2012 Border Patrol adjusted its scheduling processes to record agent time in hours rather than days and to allow agent time to be split across multiple assignments in a given day. As a result, we omitted fiscal year 2012 from the scope of our deployment scheduling data analysis.


Figure 5: Percentage of Agent Hours Scheduled for Time Off and Deployment Activities for the Southwest Border, Fiscal Years 2013–2016

Legend: Operations and Patrol = frontline activities that involve identifying and apprehending illegal entrants and identifying and seizing contraband. Some specific examples include linewatch, sign cutting, and checkpoint duties; Operational Support = activities, such as intelligence gathering or surveillance, that support frontline agents conducting operations and patrol activities; Processing = activities that occur after apprehending an individual, including transport, processing, detention, and removal; Legal Support and Litigation = activities, such as attending court proceedings, that involve prosecution of apprehended individuals; Training = activities that involve providing instruction, attending training, or completing qualification/certification tests; Administrative and Other Non-Enforcement = activities other than those above, including public relations, hiring and recruitment, and policy and compliance; Other Nonwork Time = time that agents are scheduled to be not working (other than time they are off duty) such as annual leave or scheduled sick leave.

Furthermore, it is important to note that BPETS deployment schedules reflect the scheduled availability and deployment of agents, rather than actual deployment. Actual availability or deployment may have differed because of changes in circumstances or other factors, and supervisors are not required to update BPETS to reflect these deployment changes.30 For example, an agent who was assigned to patrol the border might do so

30The data we analyzed document how agents are expected to be deployed on a given shift, at least 3 days in advance of that shift. According to Border Patrol officials, changes for an upcoming shift are normally updated and finalized just prior to the start of the shift, but changes that occur after the shift starts may not always be reflected in the scheduling data.


for part of a shift, but upon apprehending an illegal entrant the agent may spend some or all of the remainder of the shift processing the apprehended individual. According to Border Patrol officials, additional agents in the field may also be pulled from their patrol activities to conduct processing when large groups are apprehended.

Border Patrol station officials also make deployment decisions based on the unique geographical factors in their area of responsibility, such as proximity to population centers and access to certain areas (including remote areas where FOBs are located). In addition, whether the station is responsible for the operation of checkpoints is another factor that station officials consider in making decisions about how to most effectively use available agents for operations and patrol activities in the field.

Proximity to Population Centers

In relatively populated areas close to the border, the window of time Border Patrol has to respond to illegal crossings may be shorter than in more remote areas where agents may have more time to apprehend illegal crossers. Thus, proximity to population centers is a factor that officials consider when deciding how many agents to deploy to particular locations within a station’s area of responsibility. In February 2017, we reported that Border Patrol officials said that populated urban environments offer an advantageous setting for illegal entrants because within seconds to minutes these entrants can blend in with the local U.S. community after crossing the border. Therefore Border Patrol has intended to divert illicit cross-border activities into more remote or rural environments, where illegal entrants may require hours or days to reach the nearest U.S. community.31 For example, El Centro sector officials told us that an outlet mall located at the immediate border posed a threat in terms of the limited time it would take illegal crossers to assimilate into the population. Similarly, officials in one station in the Rio Grande Valley sector identified a town that is very close to the Rio Grande River, and Border Patrol agents must aim to apprehend crossers within a two to three block distance to prevent crossers from blending in with residents of 31GAO, Southwest Border Security: Additional Actions Needed to Better Assess Fencing’s Contributions to Operations and Provide Guidance for Identifying Capability Gaps, GAO-17-331 (Washington, D.C.: Feb. 16, 2017). We have also previously reported on Border Patrol’s strategy to push illegal entrants into rural areas where they can be more easily apprehended. See GAO-09-824 and GAO, INS’ Southwest Border Strategy: Resource and Impact Issues Remain After Seven Years, GAO-01-842 (Washington, D.C.: Aug. 2, 2001).

Geography





the town. Sector officials generally stated that stations prioritize deployment to areas along the immediate border.

Land and Road Access

Border Patrol may have limited access to certain areas because of land ownership or limited road infrastructure, and this may affect decisions about how to deploy agents (if at all) to these areas. Some sectors consist primarily of privately owned land, and Border Patrol officials must obtain permission from the landowner, or a judicial warrant, to access any private lands further than 25 miles from the border.32 Border Patrol officials in one sector noted that some landowners do not want Border Patrol on their property. Additionally, the availability and condition of road infrastructure can make it challenging for agents to get to some locations. For example, officials in sectors with mountainous terrain cited challenges related to accessing and patrolling mountainous areas. In particular, officials in the Tucson sector noted that the sector includes seven mountain ranges and estimated that about 20 percent of the 262 miles of land border in the sector are inaccessible by vehicle. As a result, these officials said that agents deployed to those areas patrol by foot, horseback, and air.

FOBs

In some areas where there are terrain and road access challenges, Border Patrol may establish FOBs to facilitate access to areas near the immediate border and enable agents to spend a greater proportion of their shifts on patrol. Border Patrol sector officials in the five sectors that currently operate at least one FOB, as well as officials in one other sector that previously operated a FOB, said that FOBs are beneficial for maximizing patrol time in difficult to reach locations. Assigning agents to these locations on temporary duty assignments reduces the portion of an agent’s shift that is spent in transit between the station and the patrol location. For example, officials in one sector said that the transit time between the station and the FOB is a 5 to 6 hour round trip. Thus, transit to that location could comprise 50 to 60 percent of a shift for agents 32Border Patrol agents have the authority, without a warrant, to enter private lands (but not dwellings) within 25 miles of the border to prevent the illegal entry of aliens into the United States and to search a railway car, aircraft, or vehicle for aliens within 100 miles of the border. See 8 U.S.C. § 1357(a)(3), 8 C.F.R. § 287.1(a)(2); for additional authorities, including those not specifically limited by distance from the border, see, e.g., 8 U.S.C. §§ 1225, 1357; 19 U.S.C. §§ 482, 1581, 1589a, 1595(b); 8 C.F.R. § 287.1.


deployed to that patrol area if they were to report to the station each day prior to beginning their patrol duties. Instead, agents travel between the station and the FOB only as part of the first and last day of their multi-day assignment to the FOB, and on the days in between they are pre-positioned at the FOB to begin patrols at the start of their shift.

Although FOBs can help facilitate access to some remote locations, there can also be associated challenges, and therefore, they may not be an effective solution in all cases to improving access to remote areas. For example, officials in one sector noted that Border Patrol had considered establishing a FOB to improve accessibility to that location, but there were challenges to securing the rights to access private property and providing for adequate facilities given that the area of interest did not have infrastructure to supply water to the FOB if it were to be built. In February 2016, the DHS Office of the Inspector General reported that although the challenge of supplying water to FOBs rarely causes Border Patrol to shut down a FOB, it is a frequent problem that often requires additional resources to resolve.33 Additionally, some remote or difficult to access locations may be located on private or tribal lands, which require Border Patrol to negotiate access and other aspects of FOB operations, or on wildlife refuges, which may have limitations regarding the types of infrastructure or operations in order to preserve the local habitat.

Checkpoints

Stations that have responsibility for checkpoints in their areas of responsibility consider checkpoint operations in their deployment decisions. Border Patrol’s checkpoints policy includes a recommended minimum number of agents to operate the checkpoint. The nature of a checkpoint—whether it is permanent or tactical—can also affect deployment. Permanent checkpoints are generally intended to be operational most of the time, meaning that stations with responsibility for permanent checkpoints generally assign at least the minimum number of agents to those checkpoints to ensure continuous operation.34 In contrast, tactical checkpoints are intended to be set up for short-term or intermittent use. Accordingly, a station can make a decision about whether to operate

33DHS Office of the Inspector General, OIG-16-37. 34Border Patrol policy states that checkpoint operations should be suspended if either agents or the public are in danger of being injured. Conditions that could pose such a danger include traffic congestion or weather.


a tactical checkpoint based on a determination of whether it is more effective to staff the checkpoint or whether it is more effective to deploy those agents elsewhere. Stations with responsibility for both the immediate border and interior checkpoints must balance agent deployment across both responsibilities. In contrast, border stations that do not contain checkpoints in their areas of responsibility do not have to distribute agents between checkpoint and patrol activities. Similarly, interior stations that do not have responsibility for the border can prioritize checkpoints.

Additionally, for stations with checkpoints, supervisors must determine how many agents, if any, to deploy to the areas around a checkpoint through which illegal entrants or smugglers may travel to circumvent the checkpoint (known as circumvention routes). We reported in August 2009 that Border Patrol policy highlights the need to detect and respond to circumvention activity, but at the time, officials stated that other priorities sometimes precluded positioning more than a minimum number of agents and resources in checkpoint circumvention routes.35 Similarly, as part of this review, sector officials said checkpoint circumvention routes may not be patrolled at all times because of the need to deploy agents elsewhere, including to the checkpoint itself to meet the minimum number of agents needed to keep the checkpoint operational. According to officials, in some locations, sensors and cameras assist with monitoring traffic in circumvention routes, and when technology detects traffic, agents can be deployed to respond.

In our August 2009 report, we reported that checkpoint performance can be hindered by limited staffing at checkpoints. Border Patrol policy recommended the minimum number of agents for checkpoint operation, but sector managers may have had other priorities for staff placement, and thus stations may have only staffed checkpoints—and circumvention routes—with the minimum number of agents. Additionally, as part of that review, we found that design and planning documents for the planned Interstate 19 checkpoint in the Tucson sector did not include an estimate of the number of agents who would be deployed to address circumvention activity at the new checkpoint. We recommended that, in connection with planning for new or upgraded checkpoints, CBP should conduct a workforce planning needs assessment for checkpoint staffing allocations

35GAO-09-824.



to determine the resources needed to address anticipated levels of illegal activity around the checkpoint.

In January 2017, Border Patrol began construction of a new checkpoint facility on U.S. Highway 281 south of Falfurrias, Texas, that will replace the existing checkpoint. The current checkpoint has a maximum of five lanes of traffic, whereas the new checkpoint will have a maximum of eight lanes. Border Patrol provided us an estimate for the number of agents, supervisors, and canine units that are expected to be needed to operate the new checkpoint; however, the information provided lacked supporting details, such as a discussion of what data were collected and how the data were analyzed to determine how many agents would be needed to staff the checkpoint and the surrounding circumvention routes.36 Given existing staffing constraints, having an accurate workforce planning needs assessment is important to inform future considerations for how to deploy agents to address anticipated levels of illegal activity at and around the checkpoint. Therefore, we continue to believe this recommendation is warranted.

Sector officials said they consider intelligence information—such as information about illegal traffic patterns and data on apprehensions and seizures; the types of threats in the area (e.g., illegal border crossing, drug smuggling); and transnational criminal organizations’ tactics, techniques, and procedures—when determining where to deploy available agents. Officials said they also receive information on suspected illegal traffic from community members, and stations may deploy agents to respond.

36Valid and reliable data are critical to assessing an agency’s workforce requirements as part of strategic human capital planning. For more information, see GAO, A Model of Strategic Human Capital Management, GAO-02-373SP (Washington, D.C.: Mar. 2002).

Intelligence Information about Illegal Traffic Patterns

http://www.gao.gov/products/GAO-02-373SP


From fiscal years 2013 through 2016, the nine southwest border sectors varied in how they distributed work time scheduled to activities in the six categories previously discussed—(1) operations and patrol, (2) operational support, (3) processing, (4) legal support and litigation, (5) training, and (6) administrative and other non-enforcement—although all the sectors scheduled the majority of agents’ time (between 61 and 77 percent) to operations and patrol activities.37 As shown in figure 6, the Rio Grande Valley sector scheduled the smallest percentage of agents’ work time to operations and patrol activities (61 percent) and the highest percentage of time to processing (13 percent). As discussed later in this report, the Rio Grande Valley had the highest number of apprehensions out of the nine southwest border sectors from fiscal year 2012 through 2016, thereby affecting the time needed for processing or otherwise attending to apprehended individuals.38

37These categories represent scheduled work time. In analyzing variations in deployment by sector, we focused on how agents who were scheduled to work were deployed. As such, our analysis by sector does not include comparisons of off duty time and other non-work time, such as annual leave. 38In November 2016, Border Patrol announced that it was detailing 150 agents from the Tucson, San Diego, and Del Rio sectors to processing centers in the Rio Grande Valley in light of recent increases in apprehensions along the southwest border.

Sectors Vary in How Much Time Agents Are Scheduled for Operations and Patrol Activities and Where Such Activities Are Scheduled to Occur


Figure 6: Distribution of Border Patrol Agents’ Scheduled Work Time, by Sector, Fiscal Years 2013–2016

Legend: Operations and Patrol = frontline activities that involve identifying and apprehending illegal entrants and identifying and seizing contraband. Some specific examples include linewatch, sign cutting, and checkpoint duties; Operational Support = activities, such as intelligence gathering or surveillance, that support frontline agents conducting operations and patrol activities; Processing = activities that occur after apprehending an individual, including transport, processing, detention, and removal; Legal Support and Litigation = activities, such as attending court proceedings, that involve prosecution of apprehended individuals; Training = activities that involve providing instruction, attending training, or completing qualification/certification tests; Administrative and Other Non-Enforcement = activities other than those above, including public relations, hiring and recruitment, and policy and compliance.

The scheduling data also show variations in the locations where sectors plan to deploy agents to operations and patrol activities in proximity to the border. Specifically, as shown in figure 7, the sectors ranged from scheduling 34 to 61 percent of operations and patrol time in border zones (for the Big Bend and Rio Grande Valley sectors, respectively) and from 17 to 52 percent of operations and patrol time in non-border zones (for the Del Rio and Big Bend sectors, respectively). In some cases, the data do not include a zone assignment, and sectors varied in what percentage


of operations and patrol scheduling assignments did not have a zone assignment. Including a zone assignment is not required by Border Patrol policy, and headquarters and sector officials identified some possible reasons why an assignment may not include a zone. For example, officials said that an agent could be deployed to an activity that has responsibility for multiple zones or no specific zone, such as roving patrol, specialty units (such as an intelligence unit or special operations), or assisting CBP’s Air and Marine Operations. Officials from one sector noted that a zone may not be assigned in the data because supervisors assign them orally when agents arrive at the start of a shift, and this provides supervisors flexibility to make the assignments based on the most up-to-date information about traffic patterns. Border Patrol headquarters officials said that the reasons for variations in border zone deployment are the same as we previously reported in December 2012—specifically, differences in geographical factors among the southwest border sectors (such as varying topography, ingress and egress routes, and land access issues, and structural factors such as technology and infrastructure deployments) that can affect how sectors operate and may preclude closer deployment to the border.39

39GAO-13-25.



Figure 7: Percentage of Border Patrol Agents’ Time for Operations and Patrol Scheduled for Different Zone Types, Fiscal Years 2013–2016

Note: This analysis is not comparable to a similar analysis we reported in December 2012. See GAO, Border Patrol: Key Elements of New Strategic Plan Not Yet In Place to Inform Border Security Status and Resource Needs, GAO-13-25 (Washington, D.C.: Dec. 10, 2012). For our December 2012 report, we reported on zone assignments across all agent activities, whereas the data in this report reflect zone assignments for a subset of agent time scheduled for operations and patrol activities. Additionally, in fiscal year 2012, Border Patrol changed the way it recorded scheduling data to allow supervisors to break up an agent’s workday by assigning hours to one or more activities, whereas previously supervisors could only assign a full workday to one activity.

Sectors also varied in terms of the proportion of operations and patrol time scheduled for checkpoint-related activities.40 Across the nine southwest border sectors from fiscal year 2013 through fiscal year 2016, approximately 9.4 percent of agents’ time scheduled for operations and patrol was scheduled for checkpoint activities. However, the number of agent hours scheduled for checkpoint activities—and what percentage of operations and patrol time these hours represent—vary by sector 40Border Patrol’s data include three types of activities related to checkpoints: an activity for general checkpoint responsibilities, an activity for supervisory responsibilities at checkpoints, and an activity for canine team assignments to checkpoints.



because of differences in factors, such as the number of checkpoints in a sector, the relative size of checkpoints, and the overall number of agents in a sector. For example, as shown in table 2, the El Centro and Big Bend sectors scheduled a similar number of hours to checkpoint-related activities, but these hours represented different percentages of total scheduled operations and patrol activities time—13.9 percent and 21.0 percent, respectively—which partly reflects that the El Centro sector has almost double the number of agents and fewer checkpoints than the Big Bend sector.

Table 2: Agent Hours Scheduled for Southwest Border Checkpoints, by Sector, Fiscal Years 2013–2016

Sector Number of checkpoints Number of agent hours scheduled for checkpoint activities

Number of agent hours scheduled for all

operations and patrol activities

Percent of operations and patrol scheduled

time assigned to checkpoints

Permanent Tactical

San Diego 6 20 1,182,904 12,256,126 9.7 El Centro 2 3 684,753 4,925,938 13.9 Yuma 3 8 643,517 4,314,316 14.9 Tucson 0 15 1,577,446 21,624,421 7.3 El Paso 7 9 1,688,869 13,777,824 12.3 Big Bend 4 24 683,440 3,260,394 21.0 Del Rio 5 14 497,708 8,765,982 5.7 Laredo 5 9 702,604 10,118,891 6.9 Rio Grande Valley 2 1 1,084,344 14,333,928 7.6 Total across all southwest border sectors

34 103 8,745,585 93,377,820 9.4

Source: GAO analysis of Border Patrol data. | GAO-18-50

Note: Hour amounts are rounded to the nearest whole hour.


From fiscal years 2012 through 2016, 33 percent of southwest border apprehensions were made one-half mile or less from the border, and over this time period apprehensions increasingly occurred closer to the border, as shown in figure 8. Specifically, from fiscal years 2012 through 2016, apprehensions one-half mile or less from the border increased from 24 percent to 42 percent. During the same time period, the percentage of apprehensions occurring more than 20 miles from the border steadily dropped, from 27 percent in fiscal year 2012 to 15 percent of all apprehensions in fiscal year 2016.

Apprehensions Occurred Closer to the Border in Fiscal Year 2016 Compared to Fiscal Year 2012, While Seizure Locations Remained Relatively Unchanged

Apprehensions


Figure 8: Distribution of Southwest Border Apprehensions by Distance from the Border, Fiscal Years 2012–2016

Notes: Percentages may not total to 100 because of rounding. These data exclude apprehensions with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

While all nine southwest border sectors exhibited this trend of an increase in apprehensions one-half mile or less from the border and a decrease in apprehensions farther than 20 miles from the border, the Rio Grande Valley sector had the greatest influence on the overall southwest border trend because that sector accounted for almost half (42 percent) of all southwest border apprehensions during this time period. Consistent with the overall trend for southwest border apprehensions in figure 8 above, the percentage of Rio Grande Valley sector apprehensions one-half mile or less from the border increased (from 27 percent in fiscal year 2012 to 48 percent in fiscal year 2016) and the sector’s percentage of apprehensions more than 20 miles from the border decreased (from 30 percent in fiscal year 2012 to 12 percent in fiscal year 2016). Appendix II provides more detailed information about trends in apprehensions by sector for fiscal years 2012 through 2016.

According to Border Patrol officials and apprehension data, one key driver for apprehensions occurring closer to the border is the increasing number


of apprehensions of children (either unaccompanied or as part of family units) from countries other than Mexico.41 We have previously reported that CBP officials have attributed high apprehension rates in the Rio Grande Valley sector to the high number of unaccompanied children and adults with children, many of whom turn themselves in to Border Patrol without attempting to evade detection.42 Officials said children are often told by smugglers to wait in specific locations where agents frequently patrol so that they will be found. According to Border Patrol officials, persons apprehended from Central America are often fleeing violence, and once apprehended they may assert claims for asylum in the United States.

As shown in table 3, apprehensions of individuals, particularly children, from Central American countries (specifically, El Salvador, Guatemala, and Honduras) increased, while apprehensions of Mexicans, including children, decreased. In particular, in the Rio Grande Valley sector, the number of children apprehended from El Salvador, Guatemala, and Honduras increased almost tenfold, from 6,869 in fiscal year 2012 to 60,084 in fiscal year 2016. Such apprehensions also increasingly occurred closer to the border. In fiscal year 2016, Border Patrol apprehended 36,882 children from these countries (about 61 percent) one-half mile or less from the border, compared to 1,830 (about 27 percent) in fiscal year 2012. Although other sectors accounted for smaller percentages of overall southwest border apprehensions, all sectors saw notable increases in the percent of apprehensions who were children from Central America and who were apprehended closer to the border. Border Patrol officials said other factors may also have contributed to the change in apprehension patterns, such as changes in where patrols occurred during the time period we analyzed.

41Children are those under 18 years old. Border Patrol reported that 59,692 unaccompanied alien children were apprehended in fiscal year 2016, in addition to children crossing the border as part of a family unit. Our analysis is of apprehensions where the person was under 18 and does not distinguish between unaccompanied children and those children in family units. 42GAO-17-331.



Table 3: Southwest Border and Rio Grande Valley Sector Apprehensions from Mexico and Select Central American Countries, Fiscal Years (FY) 2012 and 2016

Total Individuals (All Ages) Apprehended

Children Apprehended

FY 2012 FY 2016 FY 2012 FY 2016 All Southwest Border Sectors

All Apprehensions 355,553 405,755 30,745 101,288 From Mexico 261,366 189,267 18,918 13,731 From El Salvador, Guatemala, Honduras 86,400 197,966 11,051 84,710

Apprehensions one-half mile or less from the border

85,427 171,472 8,376 62,935

From Mexico 67,744 57,832 5,573 4,769 From El Salvador, Guatemala, Honduras 15,524 101,980 2,523 55,841

Rio Grande Valley Sector All Apprehensions 97,449 186,271 12,564 65,011


Apprehensions one-half mile or less from the border

26,110 88,993 4,040 39,235



Notes: The apprehension numbers for individuals from Mexico, El Salvador, Guatemala, and Honduras and for children reflect data for which the country of citizenship was known. For fiscal years 2012 through 2016, a total of 19 apprehension records were missing information on citizenship. We cannot determine to what extent these records may correspond to individuals from these selected countries. These data exclude apprehensions with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

From fiscal year 2012 through fiscal year 2016, seizure locations remained roughly the same, with between 64 and 70 percent of seizures occurring 10 or more miles from the border each year and between 9 percent and 11 percent of seizures occurring one-half mile or less from the border each year, as shown in figure 9.

Seizures


Figure 9: Distribution of Southwest Border Seizures by Distance from the Border, Fiscal Years 2012–2016

Notes: Percentages may not total to 100 because of rounding. These data exclude seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

Trends within individual sectors varied, but unlike with apprehensions, no single sector dominated the proportion of seizures to strongly influence the overall pattern for the southwest border. The greatest number of seizures during the 5 fiscal years occurred in the Tucson, Big Bend, and Rio Grande Valley sectors (34, 19, and 16 percent of all seizures respectively). These sectors each had different distributions of where seizures occurred, as shown in figure 10. In particular, about 1 percent of seizures in the Big Bend sector occurred within 1 mile of the border, compared to 13 percent of seizures in the Tucson sector and 37 percent of seizures in the Rio Grande Valley sector. Appendix III provides more detailed information about trends in seizures by sector for fiscal year 2012 through fiscal year 2016.


Figure 10: Percentage of Seizures by Distance From the Border for the Tucson, Big Bend, and Rio Grande Valley Sectors, Fiscal Years 2012–2016

Notes: Percentages may not total to 100 because of rounding. These data exclude seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.


According to our analysis of Border Patrol data, checkpoints accounted for about 2 percent of apprehensions and almost half of seizures in southwest border sectors. However, determining the extent to which apprehensions and seizures farther from the border are attributable to checkpoints is difficult because of data quality issues that have persisted since we previously reported on checkpoints in August 2009.43 In that report, we found that Border Patrol had established a number of measures for checkpoint performance to inform the public on program results and provide management oversight, including measures related to apprehensions and seizures at checkpoints and on circumvention routes. However, we reported that information gaps and reporting issues hindered public accountability and that inconsistent data collection and entry hindered Border Patrol’s ability to monitor the need for program improvement. Specifically, we found that a lack of management oversight and unclear checkpoint data collection guidance resulted in the overstatement of checkpoint performance results in agency performance reports, as well as inconsistent data collection practices at checkpoints. For example, officials at some checkpoints were including apprehensions that occurred within a 2.5-mile radius of the checkpoints in their reporting of apprehensions at checkpoints, which led to inconsistent reporting across checkpoints. We reported that the lack of oversight and unclear data collection guidance hindered management’s ability to monitor the need for program improvement. We therefore recommended, among other things, that Border Patrol establish internal controls for

43GAO-09-824.

Long-Standing Data Quality Issues Make It Difficult to Precisely Measure Checkpoints’ Contributions to Apprehensions and Seizures Checkpoints’ Role in Apprehensions and Seizures Is Difficult to Measure with Precision Because Of Long-Standing Data Quality Issues



management oversight of the accuracy, consistency, and completeness of checkpoint performance data.44

In response to our recommendations, Border Patrol issued several memoranda in 2009 and 2010 related to the collection of checkpoint data, including guidance intended to distinguish between apprehensions and seizures occurring at checkpoints compared to those occurring in circumvention routes.45 In particular, these memoranda stated that:

• “At the checkpoint” is defined as the area including the checkpoint itself and the roadway prior to the checkpoint marked with cones and/or warning signs related to checkpoint operations (which, according to Border Patrol’s checkpoint policy, are to begin on the roadway one-half mile from the checkpoint itself).46 Apprehensions and seizures occurring at a checkpoint are to be recorded by selecting the appropriate checkpoint location from a dropdown list of landmarks (landmark data field).47

• “Circumvention” is defined as “any deviation from a normally used route of egress in order to avoid detection by a checkpoint,” and if an individual was apprehended while attempting to circumvent a checkpoint, the apprehension is to be recorded by marking a checkbox labeled “Circumvention App?” (There is no data field for seizures that indicates that Border Patrol seized contraband from someone attempting to circumvent a checkpoint, but the seizure can be associated with an apprehension or arrest record for the person carrying the contraband,

44As we reported in 2009, internal control standards require that agencies accurately record and report data necessary to demonstrate agency performance, and that they provide proper oversight of these activities. See GAO, Standards for Internal Control in the Federal Government, GAO/AIMD-00-21.3.1 (Washington, D.C.: Nov. 1999). 45The memorandum that defined how data for checkpoint and circumvention apprehension should be recorded was issued in October 2009. The memorandum that defined an occurrence at a checkpoint compared to a circumvention was issued in January 2010. In addition, an April 2016 memorandum reminded sector officials of the importance of recording accurate and consistent data. 46Based on this definition, we considered an apprehension to occur “at the checkpoint” if it occurred one-half mile or less from the checkpoint. 47Sector officials define landmarks and which landmarks are checkpoints within e3 for their respective sectors. Border Patrol began recording GPS coordinates for apprehensions and seizures in 2011, and prior to that, landmarks were used to denote the general area where an apprehension occurred.

http://www.gao.gov/products/GAO/AIMD-00-21.3.1


and the apprehension or arrest record may have the “Circumvention App?” box checked.)

However, as discussed below, these memoranda have not fully addressed our recommendation because our analysis indicates that issues persist regarding the accuracy and consistency of data on checkpoint apprehensions and seizures. These issues continue to affect how Border Patrol monitors and reports on checkpoint performance results.

According to Border Patrol officials, since the implementation of these memoranda, Border Patrol has reported on apprehensions and seizures at checkpoints based solely on the landmark data field. Specifically, an apprehension or seizure event is reported as having occurred at a checkpoint if the landmark associated with the event corresponds to the landmark for a checkpoint (checkpoint landmark). In September 2016, the Border Patrol Chief testified before a congressional committee that Border Patrol apprehended 8,503 individuals and seized over 75,000 pounds of drugs at checkpoints nationwide in fiscal year 2015, and the officials responsible for overseeing and analyzing the data said that these numbers were generated by determining the number of apprehensions and seizures associated with a checkpoint landmark.48 Furthermore, CBP’s fiscal year 2018 congressional budget justification noted that measurement of checkpoint activities—such as apprehensions at checkpoints—can gauge checkpoint operational effectiveness and provide insight into the effectiveness of the Border Patrol’s overall national border enforcement strategy. CBP reported in the budget justification that apprehensions at checkpoints ranged from 1.34 to 2.52 percent of nationwide apprehensions across fiscal years 2013 through 2016.

To assess Border Patrol’s efforts to implement our August 2009 recommendation and determine the extent to which Border Patrol’s reporting of checkpoint statistics provides accurate information about enforcement actions at and around checkpoints, we analyzed

48Mark A. Morgan, Chief, U.S. Border Patrol, Moving the Line of Scrimmage: Re-examining the Defense-in-Depth Strategy, testimony before the U.S. House of Representatives, Committee on Homeland Security, Subcommittee on Border and Maritime Security, 114th Cong., 2nd sess., September 13, 2016.


apprehension and seizure data from fiscal years 2013 through 2016.49 For example, as shown in table 4, an apprehension or seizure event that occurred one-half mile or less from a checkpoint (according to the GPS coordinates of the event) and that was also associated with the nearest checkpoint landmark was considered category 1.

Table 4: Categories Describing the Relationship between Apprehension and Seizure (Event) Locations and Checkpoints

Category Global Positioning System (GPS) coordinates

Landmark associated with the event

Included in Border Patrol reporting of events at checkpoints

Implications for analysis and reporting of checkpoint activities

1 One-half mile or less from a checkpoint

Corresponds to the nearest checkpoint landmark

Yes Event occurred at a checkpoint—both the GPS coordinates and landmark indicate that the event occurred at a checkpoint.

2 One-half mile or less from a checkpoint

Corresponds to a location other than the nearest checkpoint landmark

No Event’s relation to a checkpoint is unclear—the GPS coordinates indicate that the apprehension or seizure occurred one-half mile or less from a checkpoint, but the landmark indicates that the apprehension or seizure did not occur at a checkpoint.

3 Farther than one-half mile from a checkpoint

Corresponds to the nearest checkpoint landmark

Yes Event’s relation to a checkpoint is unclear—the GPS coordinates indicate that the apprehension did not occur at a checkpoint, but Border Patrol’s reporting of apprehensions and seizures at checkpoints would consider these as occurring at checkpoints because of the landmark.

4 Farther than one-half mile from a checkpoint

Corresponds to a location other than the nearest checkpoint landmark

No Event did not occur at a checkpoint—both the GPS coordinates and landmark indicate that the apprehension or seizure did not occur at a checkpoint.

Source: GAO analysis of Border Patrol data and policies. | GAO-18-50

Note: We based these categories on Border Patrol’s policy that “at a checkpoint” refers to the area including the checkpoint itself and the roadway prior to the checkpoint marked with cones and/or warning signs, which are to begin one-half mile from the checkpoint.

49Specifically, for each apprehension and seizure record for this time period, we analyzed the landmark associated with the apprehension or seizure and the GPS coordinates. We also analyzed the use of the “Circumvention App?” field to determine the extent to which this field could be used to provide information about other effects of checkpoints. We assigned each apprehension and seizure (event) into one of four categories based on whether the GPS coordinates for the event occurred close enough to the GPS coordinates for a checkpoint to be considered “at a checkpoint” and whether the event’s landmark corresponds to the nearest checkpoint landmark. Our analysis of checkpoint apprehensions is limited to fiscal years 2013 through 2016 because Border Patrol began maintaining checkpoint lists for such historical analysis in fiscal year 2013.


Our analysis of Border Patrol data, as shown in table 5, indicates that at least 31,639 apprehensions and 30,449 seizures—those that are in category 1—occurred at checkpoints from fiscal years 2013 through 2016 based on both the GPS coordinates and the landmarks associated with those apprehensions and seizures.50 These apprehension and seizure events would be considered as occurring “at checkpoint” for Border Patrol reporting purposes because a checkpoint landmark was associated with the event.

However, for the 19,759 apprehensions and 1,182 seizures in category 2—which are not included in Border Patrol’s reporting—it is unknown what proportion should be considered “at a checkpoint.” This is because for each of these apprehensions and seizures, the associated landmark does not correspond to the nearest checkpoint landmark, even though the GPS coordinates indicate that these apprehensions and seizures occurred one-half mile or less from a checkpoint location. Border Patrol officials said that one reason why the checkpoint landmark might not be indicated for apprehensions and seizures that occur one-half mile or less from a checkpoint is if the checkpoint is nonoperational at the time. However, our analysis suggests that not all apprehensions and seizures recorded in category 2 would reflect instances of checkpoints being non-operational. For example, about 30 percent of apprehensions that were one-half mile or less from the Falfurrias, TX, checkpoint (4,278 of 14,345 apprehensions) did not use the landmark for that checkpoint. Border Patrol officials in the Rio Grande Valley sector said the Falfurrias checkpoint is rarely closed, so the checkpoint being closed does not fully explain why the relevant checkpoint landmark was not used. Because Border Patrol’s policies do not provide guidance about recording data differently when a checkpoint is operational or nonoperational, it is unclear what proportion of apprehensions or seizures in category 2 reflect inconsistent application of Border Patrol’s guidance versus instances of a checkpoint being nonoperational.

50Sectors varied in the extent to which their apprehensions occurred at checkpoints. See appendix II for our checkpoint analysis of apprehensions by sector.


Table 5: Southwest Border Checkpoint Apprehensions and Seizures by Location Category, Fiscal Years 2013–2016

Location Category Number of apprehensions

(percent of total southwest border

apprehensionsa)

Number of apprehensions with

“Circumvention Apprehension?” box

checkedb

Number of seizures (percent of total

southwest border seizuresa)

1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

31,639 (1.9 percent)

165 30,449 (43.3 percent)

2. Occurred one-half mile or less from the checkpoint and is associated with a landmark for a location other than the nearest checkpoint

19,759 (1.2 percent)

592 1,182 (1.7 percent)

3. Occurred farther than one-half mile from the checkpoint and is associated with the landmark for the nearest checkpoint

1,746 (0.1 percent)

108 582 (0.8 percent)

4. Occurred farther than one-half mile from the checkpoint and is associated with a landmark for a location other than the nearest checkpoint

1,570,437 (96.7 percent)

27,682

38,123 (54.2 percent)

Total 1,623,581 28,547 70,336


Note: These data exclude apprehensions and seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors. aPercentages may not total to 100 because of rounding. bThe number of apprehensions where the “Circumvention Apprehension?” box is checked is a subset within the overall number of apprehensions for each location category.

There are also inconsistencies in how Border Patrol is recording and reporting on apprehensions and seizures on potential circumvention routes. Events in category 3 appear to have occurred in circumvention routes rather than at checkpoints—they occurred farther than one-half mile from a checkpoint, and thus do not fit Border Patrol’s definition of an apprehension that occurs “at a checkpoint”—but because they are associated with a checkpoint landmark, Border Patrol’s reporting of events at checkpoints includes these apprehensions and seizures.

Additionally, officials responsible for compiling checkpoint data said that they have not analyzed the use of the “Circumvention App?” checkbox to separately determine apprehensions that occur around checkpoints. Although the GPS coordinates and associated landmarks suggest that apprehensions in category 4 are not related to checkpoints, there were over 27,000 apprehensions in this category that had the “Circumvention App?” box checked. However, these apprehensions have not been included in statistics related to checkpoints because Border Patrol’s


reporting to date has focused on events associated with checkpoint landmarks and has not separately analyzed or reported the number of apprehensions for which the “Circumvention App?” box was checked. In doing so, Border Patrol’s reporting does not differentiate between apprehensions that occurred at versus around a checkpoint. Border Patrol officials agreed that the agency’s policies could better differentiate between these areas and how to record data for events that occur in each location.

Examining apprehensions specific to an individual checkpoint further illustrates the inconsistencies in data recorded for checkpoints. Figure 11 shows how apprehensions at and around one checkpoint have been recorded using GPS coordinates and landmarks, in relation to the one-half mile radius around the checkpoint.

Figure 11: Maps of Apprehensions Surrounding Two Checkpoints, Fiscal Years 2013–2016

Notes: Not all apprehensions attributable to these checkpoints are visible on the maps due to multiple apprehensions occurring at the same location and apprehensions occurring beyond the scale of the


map. In particular, some of the apprehensions farther than one-half mile from the checkpoint and for which the landmark code matches the checkpoint are outside the area depicted. These data exclude apprehensions with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

Border Patrol’s methodology for determining the number of apprehensions and seizures at checkpoints—which counts only apprehensions and seizures associated with checkpoint landmarks—may result in overstating or understating apprehensions and seizures that occurred at checkpoints; however, the precise number of apprehensions and seizures that occurred at checkpoints cannot be determined because of the data inconsistencies noted above. For example, Border Patrol’s reporting—such as in the Border Patrol Chief’s testimony or CBP’s fiscal year 2018 budget justification—may overstate apprehensions at checkpoints by including apprehensions in category 3, while it may understate apprehensions by not including some portion, or all, of the apprehensions in category 2. For the 4 fiscal years of data we analyzed, this means that Border Patrol’s methodology for attributing apprehensions to checkpoints would potentially overstate by 1,746 apprehensions (about 0.1 percent of total southwest border apprehensions) and potentially understate by as many as 19,759 apprehensions (about 1.2 percent of total southwest border apprehensions). Although these numbers represent relatively small percentages of total southwest border apprehensions, they are important for the measurement of checkpoint apprehensions given that Border Patrol has generally reported that about 2 percent of apprehensions occur at checkpoints, and in particular, adding 1.2 percentage points to the reported 2 percent would increase the reported contributions of checkpoints by about 50 percent.

Although Border Patrol issued guidance in 2009 and 2010 in response to our recommendation, our analysis demonstrates that this guidance does not provide sufficient clarity on how data are to be recorded, and as a result data quality issues have persisted. For example, Border Patrol’s guidance does not indicate what landmark should be used when an agent apprehends an individual who was attempting to circumvent a checkpoint. Additionally, Border Patrol has not provided sufficient oversight of the accuracy, consistency and completeness of checkpoint data since the guidance was issued. In July 2013, Border Patrol issued a memorandum to establish the Checkpoint Program Management Office (CPMO), and the memorandum tasked CPMO with overseeing checkpoint data quality and accuracy, among other things. However, CPMO was not officially formed until the summer of 2016 when we began this review. Officials noted that while Border Patrol staff had been consistently assigned to


oversee checkpoint data as a collateral duty, these assignments were not within an officially formed CPMO and there was no centralized oversight of checkpoint data or performance. The Associate Chief responsible for overseeing CPMO told us he had not been aware of the memorandum establishing CPMO until we requested checkpoint policies as part of this review, and he explained that CPMO had not been formally established under his predecessor at the time of the July 2013 memorandum.

In late summer 2016, the Associate Chief formally established CPMO with the two Border Patrol agents who were, at the time, assigned part-time to oversee checkpoints. However, the CPMO establishing memo called for two full-time staff members, and one of the staff assigned to CPMO part-time moved to another position within Border Patrol several months later. The first full-time staff person was assigned to CPMO in January 2017. In March 2017, CPMO officials said they agreed with our findings regarding inconsistent recording of checkpoint data, and they said they have drafted a policy to provide additional guidance, including how to distinguish how data are recorded for apprehensions and seizures that occur at the checkpoint versus around the checkpoint. The Assistant Chief for CPMO, in consultation with sector and data analysis officials, has drafted additional guidance for recording apprehensions and seizures data in a manner that differentiates between events that occurred at versus around checkpoints. According to this official, this guidance will be included in a larger update to Border Patrol’s checkpoint policy because the checkpoint policy was last updated in 2003. Border Patrol officials said they expect the updated checkpoint policy with additional data entry guidance and procedures will be in place by March 2018, following Border Patrol and CBP management review and approval and programming changes to Border Patrol’s data systems.

Having quality control procedures in place to accurately document apprehensions and seizures that occur at and around checkpoints is important to enable Border Patrol to measure checkpoint effectiveness and to make better deployment decisions about the extent to which circumvention routes should be staffed. Distinguishing between the locations of apprehensions and seizures, relative to checkpoints, would provide more visibility into illegal traffic patterns at and around checkpoints that can be used for staffing and other resource decisions. Until revised internal control practices are in place, including data collection guidance and sufficient oversight of the recording of the data, our 2009 recommendation that Border Patrol establish internal controls for management oversight of the accuracy, consistency, and completeness of checkpoint performance data remains warranted. As


part of our regular follow up on implementation of our recommendations, we will monitor Border Patrol’s progress in issuing and implementing the planned update to its checkpoint policy.

In addition to analyzing where apprehensions and seizures occurred, we analyzed marijuana seizure data to determine how seizures that occurred at checkpoints compared to those that occurred at other locations.51 As shown in figure 12, out of the 30,449 seizures that occurred at checkpoints, at least 12,214 (40 percent) were 1 ounce or less of marijuana52 seized from U.S. citizens.53 In contrast, seizures occurring at non-checkpoint locations were more often higher-quantities seized from aliens. For example, more than three-quarters of marijuana seizures at non-checkpoint locations were of over 50 pounds (25,792 out of 33,477

51We analyzed seizure data using the four location categories previously identified (based on GPS coordinates and landmarks), and because it is unknown the extent to which seizures in categories 2 and 3 are related to checkpoints, we focused on comparing category 1 seizures against category 4 seizures. Seizures in categories 2 and 3 represent a small percentage (1.7 percent and 0.8 percent, respectively) of seizures over the time period we analyzed. Furthermore, because the majority (about 92 percent) of all seizures from fiscal years 2013 through 2016 consisted of narcotics, and most (86 percent) of those narcotics seizures were marijuana, we compared marijuana seizures at checkpoints against marijuana seizures at other locations (non-checkpoint seizures). Our analysis of checkpoint seizures compared to other seizures is limited to fiscal years 2013 through 2016 because Border Patrol began maintaining checkpoint lists for such historical analysis in fiscal year 2013. In comparing checkpoint and non-checkpoint seizures, we analyzed the quantities seized and the reported alienage or citizenship status, and removability, of the individuals arrested in connection with the seizure—in other words, whether the individuals were U.S. citizens, potentially removable aliens, or not subject to removal. See appendix III for the number of checkpoint seizures for each sector during the 4 years analyzed. 52Under federal law, anyone who is found in possession of marijuana in any amount may be subject to criminal prosecution. Possession with intent to manufacture, distribute or dispense a controlled substance carries more severe criminal penalties (21 U.S.C. § 841) in comparison to simple possession (21 U.S.C. § 844). Possession of a personal use amount of certain controlled substances, including marijuana (1 ounce or less), would generally subject an individual to a civil penalty. See 21 U.S.C. § 844a. 53As noted in figure 12, some seizures are not associated with a particular arrest or apprehension record, and in these cases we cannot determine the alienage or citizenship of the individual arrested in connection with the seizure. In particular, there were 2,952 marijuana seizures at checkpoints not associated with a particular arrest or apprehension record, and 1,678 of these were for quantities of 1 ounce or less.

Almost Half of Seizures that Occurred at Checkpoints Were One Ounce or Less of Marijuana from U.S. Citizens


seizures).54 (Appendix III includes additional detail on the distribution of marijuana seizures by quantity seized.)

54The majority of marijuana seizures at non-checkpoint locations (63 percent, or 21,205 out of 33,477) were not associated with a particular arrest or apprehension record, and in these cases we cannot determine the alienage or citizenship of the individual arrested in connection with the seizure. Out of the 33,477 marijuana seizures at non-checkpoint locations, 7,583 were seized from a potentially removable alien, 2,588 were seized from a non-removable alien, 2,101 were from U.S. citizens, and the remaining 21,205 were not associated with a particular arrest or apprehension record.


Figure 12: Breakdown of Seizures by Type of Seizure, Location of Seizure, and Individuals Associated with the Seizure, Fiscal Years 2013–2016

Notes: These data exclude apprehensions with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors. aThe category for unspecified subject refers to a seizure for which a Border Patrol agent did not attribute the seizure to a particular individual who was arrested or apprehended in conjunction with


the seizure. Some examples when this might occur are if there were multiple people in the vehicle that contained the contraband or if Border Patrol found the contraband while it was unattended. bAs discussed in this report, the relationship between some seizures and checkpoints is unclear because such seizures either (a) occurred one-half mile or less from the checkpoint and were associated with a location other than the nearest checkpoint or (b) occurred farther than one-half mile from the checkpoint but were associated with the checkpoint location.

Border Patrol officials said that the primary purpose of checkpoints is to enforce immigration laws, but agents at checkpoints are also expected to take action when they incidentally encounter violations of other federal laws. In particular, they noted that when a trained canine alerts agents to the presence of a concealed human or substance the canine was trained to detect, agents are required to respond to the alert. Based on the canine alert, agents do not know until they conduct a search of the vehicle what the canine detected (concealed human or illicit substance) or what quantity of a substance might be present—and therefore, agents cannot determine prior to an inspection whether the occupants of the vehicle are travelling with what would generally be considered a personal use quantity of a substance or whether they are carrying larger quantities potentially with the intent to distribute, dispense, or manufacture.


Members of state and local law enforcement and business and community groups that we spoke to generally support Border Patrol’s efforts, but some raised concerns about checkpoint operations and the broader defense in depth strategy. Members of all three community groups we met with during our visits to the Rio Grande Valley and Tucson sectors generally supported Border Patrol. Additionally, officials from law enforcement agencies we interviewed generally said they had a positive working relationship with Border Patrol and that Border Patrol has played a role in limiting cross-border illicit activity in their communities. For example, one law enforcement official from the Tucson sector said that the community would be overwhelmed without Border Patrol’s efforts in the area, and another said that without the defense in depth approach, illegal activity would likely be worse, although this latter official noted there can be communication and coordination challenges in working with Border Patrol. Some residents and law enforcement officials we met with in the two sectors we visited said that they support Border Patrol’s use of checkpoints. For example, the leader of one community group said the group’s members viewed checkpoints positively, and members from another group said that some residents in their community believe that their local checkpoint is making the community safer through law enforcement presence.

However, Border Patrol’s defense in depth deployment strategy may also result in communities ranging up to 100 miles from the border experiencing effects associated with Border Patrol enforcement actions to interdict illicit cross-border activity. In April 2015, we reported that illicit cross-border activity can negatively affect business and the safety of

Collecting Additional Data Could Improve Border Patrol’s Existing Efforts to Identify and Respond to Community Concerns Regarding the Defense in Depth Strategy

Defense in Depth Strategy’s Effects on Surrounding Communities are Difficult to Quantify, but Collecting Additional Data Could Inform Actions to Address Such Effects


farms and ranches on or near the border.55 Although data are limited to support the extent of criminal activity tied to cross-border illegal traffic, available data indicate that cross-border traffic affects areas beyond the immediate border. For example, in fiscal year 2016, 20 percent of all Border Patrol apprehensions and 77 percent of all seizures occurred more than five miles from the border. Therefore, illegal crossers and drug smugglers may sometimes travel near or through communities and private property in areas that are not along the immediate the border, prior to being apprehended by Border Patrol. For example, members of one community group we interviewed said that there are hundreds of illegal crossers and smugglers who attempt to circumvent the local checkpoint by walking through the surrounding ranches. Echoing views from ranchers we interviewed for a December 2012 report,56 members of one community group we spoke with as part of this review said that they would like to see Border Patrol direct more enforcement efforts at the immediate border to prevent illegal crossers from entering their communities or properties. Officials we interviewed from two sheriffs’ departments in nearby counties said they have heard similar views from residents.

Community groups and law enforcement officials we met with as part of this review identified concerns regarding private property damage and public safety resulting from illegal cross border traffic, similar to concerns we have reported in the past.

• Private Property Damage: Community members have reported damage to private property suspected to have occurred as a result of individuals trying to illegally cross the border or Border Patrol enforcement actions. Border Patrol officials we spoke with in six of nine sectors cited concerns from community residents about illegal crossers and Border Patrol agents traveling on their private property. Additionally, officials from two sheriffs’ departments told us that ranchers in their communities have voiced complaints about damage on their properties resulting from illegal crossers or Border Patrol activity. These concerns are similar to concerns we identified in an April 2015 report, in which we reported that landowners had reported damage to private property—including broken

55GAO-15-399. 56GAO-13-25.




gates, destroyed crops, and injured or lost livestock—as a result of individuals trying to illegally cross the border (see fig. 13).57

Figure 13: Mangled Gate Suspected to Have Been Caused by Illegal Border Crossers along Highway 281 in Brooks County, Texas

In addition to identifying damage suspected to be caused by illegal crossers, landowners we spoke with as part of that review also reported damage that may have resulted from Border Patrol’s enforcement efforts. We previously reported in April 2015 that some landowners had filed tort claims alleging damage to their property as a result of the conduct of an employee of Border Patrol or any CBP component that was acting within the scope of his or her official duties. Examples of such claims include CBP vehicles crashing through properties and damaging fences, gates, irrigation pipes, and crops.

57GAO-15-399.



• Public Safety: Additionally, according to Border Patrol and local law enforcement officials, illegal entrants and smugglers could pose a public safety risk to communities along the border or further inland. We previously reported in December 2012 that ranchers in the Tucson sector said they were most concerned about safety.58 Officials from law enforcement agencies that we interviewed as part of this current review said that crime resulting from illicit cross-border activity has affected border communities. In particular, law enforcement officials we spoke with cited drug smuggling (including recruiting juveniles to engage in drug smuggling), home invasions, burglaries, and vandalism. The effects related to public safety and private property associated with Border Patrol’s defense in depth strategy may be felt more acutely in communities near checkpoints; in particular, one of Border Patrol’s stated goals for checkpoints is to deter and disrupt smuggling efforts, and as a result, smuggling traffic may be pushed onto checkpoint circumvention routes, which may pass through these communities. We previously reported in August 2009 that Border Patrol officials acknowledge that this approach can adversely impact communities near checkpoints, and said that sometimes there were not enough agents in place to deter illegal activity or apprehend trespassers in surrounding areas.59 As noted earlier in this report, this remains true—checkpoint circumvention routes are not always patrolled.

We are unable to measure the extent Border Patrol’s defense in depth strategy has affected communities through measures such as crime rates or effects on property values. As part of previous reviews, we have reviewed information related to the impacts of illegal cross-border activity on local communities, including reports of property damage (such as tort claims) and available crime data. As a result, we have previously reported that methodological challenges existed and data were unavailable to substantiate the extent to which illegal border crossings and drug smuggling have affected local communities in terms of public safety and private property damage. In August 2009 we reported that a comparison of community impacts for the time before and after a checkpoint was established would require a complete set of historical data to develop a baseline understanding, before interpreting factors that can change the baseline. However, there are limited data sets for specific geographic areas around checkpoints, with county level data being the smallest possible geographic area, in many cases. For instance, in terms of crime 58GAO-13-25. 59GAO-09-824.




data, officials from one police department in the Tucson sector told us that they did not track criminal activity committed by illegal entrants. In 2011, as part of Border Patrol’s efforts to implement our August 2009 recommendations, Border Patrol requested a study to identify the effects of checkpoints on nearby communities and develop an approach to measure these effects, and this study also noted data limitations that affect conclusions regarding the effects of checkpoints on surrounding communities.60

Implementing two of our August 2009 recommendations could help Border Patrol collect relevant data to examine the community effects of checkpoint operations specifically and take corresponding actions to respond to ongoing community concerns. In August 2009 we reported that Border Patrol had previously identified performance measures to examine the effect checkpoint operations have on quality of life in the surrounding communities, but the agency was not using these measures. As a result, Border Patrol was hindered in its ability to assess the impact of checkpoints on local communities. We recommended that Border Patrol (1) implement quality of life measures identified by Border Patrol to evaluate the impact that checkpoints have on local communities; and (2) use the information generated from the quality of life measures in conjunction with other relevant factors to inform resource allocations and address identified impacts.

Border Patrol agreed with the recommendations but has not yet fully implemented them. In 2010, Border Patrol asked a DHS Center of Excellence, co-led by the University of Arizona and the University of Texas at El Paso, to conduct a study to help address our recommendations. The resulting December 2012 report made several recommendations to Border Patrol on evaluating the impact of

60In 2014, at the request of Border Patrol, the National Center for Border Security and Immigration at the University of Arizona released a study that investigated the effect that the construction of a canopy on a checkpoint in the Tucson sector had on real estate values. A regression analysis provided marginally statistically significant evidence of downward pressure on housing prices south of the checkpoint following the canopy’s construction compared to housing prices north of the checkpoint. The analysis was focused on the difference in real estate prices in one geographic area before and after the construction of a canopy at one checkpoint, therefore any suggested effects from this one case study is not generalizable to apply to other checkpoints along the southwest border. Moreover, according to a written statement submitted to Congress by the study’s co-author, because of limited data availability and the difficulty in isolating the checkpoint canopy’s effect from those of the housing crisis and other economic conditions, the results should be seen as suggestive rather than definitive.


checkpoints on local communities using quantitative measures and with maintaining regular contact with the public to elicit opinions on experiences with the checkpoint, both positive and negative. Border Patrol has since reported plans for implementing our recommendations but has revised the estimated completion dates several times. (See appendix I for more information about Border Patrol’s planned actions to address these recommendations.) As discussed later in this report, Border Patrol provides opportunities for members of the community to express concerns related to the defense in depth strategy since our previous review of checkpoint operations in 2009, however, some residents and local law enforcement officials near checkpoints we spoke to for this review remain concerned about the effects checkpoints may have on their communities. Measuring performance, such as quality of life measures related to checkpoints, would give Border Patrol critical information on which to base decisions for improving checkpoint operations. Therefore, we continue to believe that our recommendations remain warranted.

Border Patrol uses a variety of methods to collect feedback from community members related to the defense in depth strategy. It receives feedback through direct communication and informal relationships, which are facilitated in part by communication and outreach events organized by sector Border Community Liaison (BCL) programs. Border Patrol initiated the BCL program in April 2011 in an effort to enhance Border Patrol’s relationships with landowners and the community as a whole. According to the July 2012 CBP implementation memo, the BCL program’s function and associated positions are intended to enhance CBP’s interaction with communities and provide a fact-based understanding of community views, concerns, and issues as they relate to CBP. According to Border Patrol officials, sector BCL agents interact with members of the local community to address complaints and also introduce the community to how Border Patrol operates so that there is a better understanding and relationship between Border Patrol agents and the surrounding community.

Each sector has its own BCL program designed to address complaints and improve the relationship between Border Patrol agents and the surrounding community, and the efforts within each program range from official events to informal communications. Sector and station BCL programs organize official events such as cook-offs, stakeholder events, and open houses where community members learn about Border Patrol’s activities and have the opportunity to share their concerns. As an

Border Patrol Is Taking Steps to Identify and Respond to Community Concerns


example of informal communications, Border Patrol officials from one station in the Rio Grande Valley sector told us that agents and officials make an effort to be very approachable to community members, as demonstrated through actions such as the station’s patrol agent in charge providing a personal cell phone number to local residents to facilitate direct communication.

In addition, every southwest border sector uses the Compliments and Complaints Management System (CCMS). The CCMS is a computerized system that allows users to log and track complaints or compliments. The CCMS is meant to identify trends and patterns in community comments to better address complaints and compliments, but Border Patrol officials have questioned its usefulness. Following a pilot program, in January 2017, CCMS became a permanent program to all CBP offices that have interaction with the public. Comments can be entered directly by residents or by Border Patrol officials who have received feedback from the community. According to the memo, CBP also standardized the response time for compliments and complaints entered into the system throughout the agency. Agency officials are to send an acknowledgment of receipt within 5 business days and complete responses within 45 days.

Officials from six of nine sectors said they generally preferred the less formal methods of interacting with the community, as discussed above, compared to the CCMS. Some of the reasons they identified included that community members often prefer to speak with an agent instead of inputting their concern into a system, very few complaints or compliments are logged into the CCMS by residents, the system is not user friendly, and it is rarely used for data recall. According to a report generated by Border Patrol headquarters, there were 599 comments entered into the CCMS nationwide in calendar year 2016. Of those, 81 were compliments.

Border Patrol takes various actions to respond to community concerns it has identified, including considering the input of local stakeholders when making deployment decisions. For example, officials from the Tucson sector told us that agents engage with ranchers who have game cameras on their properties so station officials can consider the flow of illegal entrants or drug smugglers on their properties when making deployment decisions. Moreover, officials from the Rio Grande Valley sector said that sector and station officials take into account population centers when making deployment decisions to attempt to deploy agents in positions to apprehend entrants prior to reaching population centers because once they enter the general population they are more difficult to detect and apprehend. Additionally, community members and Border Patrol officials


told us that agents respond to calls of suspected illegal cross-border activity on private lands. Border Patrol has various mechanisms in place for community members to notify agents of suspected activity. For example, one station in the Rio Grande Valley sector created a mobile phone application and released a limited number of licenses for ranchers and landowners to take a picture if they see suspicious activity and send it directly to Border Patrol. Moreover, landowners in the Rio Grande Valley sector told us that Border Patrol has been responsive to calls when something out of the ordinary has been spotted on private land.

We provided a draft of this report to the Department of Homeland Security for their review and comment. In its comments, reproduced in appendix IV, DHS provided an update on planned actions to implement the four open recommendations from our August 2009 report. DHS also provided technical comments, which we incorporated as appropriate.

We are sending copies of this report to the appropriate congressional committees and the Secretary of Homeland Security. In addition, the report is available at no charge on the GAO website at http://www.gao.gov.

If you or your staff have any questions about this report, please contact me at (202) 512-8777 or [email protected]. Contact points for our Offices of Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix V.

Rebecca Gambler Director, Homeland Security and Justice Issues

Agency Comments

http://www.gao.gov/


Appendix I: Status of GAO Recommendations Related to Checkpoints

Page 57 GAO-18-50 Border Patrol

In August 2009, we reported on and made recommendations regarding the measurement of checkpoint performance and the impact of checkpoint operations on nearby communities, among other things related to checkpoints.1 In comments provided on our August 2009 report, the Department of Homeland Security (DHS) concurred with those recommendations. This appendix provides additional detail regarding the status of the recommendations from that report, including two recommendations that U.S. Customs and Border Protection (CBP) has implemented.

Recommendation 1: Establish milestones for determining the feasibility of a checkpoint performance model that would allow the Border Patrol to compare apprehensions and seizures to the level of illegal activity passing through the checkpoint undetected.

Status: Closed – Implemented

In August 2009, we reported that the Border Patrol had developed some useful measures of checkpoint performance, but the agency lacked a model or method that would allow the agency to compare the number of apprehensions and seizures made at the checkpoint to the level of illegal activity passing through the checkpoint undetected. The lack of this information challenged the Border Patrol’s ability to measure checkpoint effectiveness and provide public accountability. In 2010, Border Patrol asked a DHS Center of Excellence to study checkpoint performance, including developing a checkpoint performance model, and the DHS Center of Excellence issued its report in December 2012. In June 2013, Border Patrol reported that the agency had considered the checkpoint performance models proposed by the National Center for Border Security and Immigration—the DHS Center of Excellence—but determined it was not feasible to use the proposed models due to cost prohibitions and other factors. This action was responsive to the intent of our recommendation to study the feasibility of a checkpoint performance model, and this recommendation has been closed as implemented.

Recommendation 2: Establish internal controls for management oversight of the accuracy, consistency, and completeness of checkpoint performance data. 1GAO, Border Patrol: Checkpoints Contribute to Border Patrol’s Mission, but More Consistent Data Collection and Performance Measurement Could Improve Effectiveness, GAO-09-824 (Washington, D.C.: Aug. 31, 2009).





Status: Open

In August 2009, we reported on inconsistencies in the way agents collected and entered performance data into the checkpoint information system. As a result, data reported in the system were unreliable. In October 2009, Border Patrol issued a memorandum specifying which data fields agents should use to indicate that an enforcement activity occurred at a checkpoint (or on a circumvention route, for apprehensions), and in January 2010 Border Patrol issued an additional memorandum on checkpoint data integrity that further specified definitions for “at the checkpoint” and “circumvention.” In subsequent years, Border Patrol officials reported to us that they were taking steps to develop a redesigned checkpoint information system, implement a data oversight procedure, and provide training, and estimated completion dates were revised several times. In its comments on this report (see app. IV), DHS stated that it expects to issue an updated checkpoint policy, including updates on data entry guidance and oversight to address data integrity, by February 28, 2018. As discussed earlier in this report, data quality issues have persisted, and without established internal controls, the integrity of Border Patrol’s performance and accountability system with regard to checkpoint operations remains uncertain.

Recommendation 3: Implement the quality of life measures that have already been identified by the Border Patrol to evaluate the impact that checkpoints have on local communities. Implementing these measures would include identifying appropriate data sources available at the local, state, or federal level, and developing guidance for how data should be collected and used in support of these measures.

Status: Open

In August 2009, we reported that Border Patrol had identified some measures to evaluate the impact that checkpoints have on local communities in terms of quality of life, but Border Patrol had not implemented the measures. As a result, the Border Patrol lacked information on how checkpoint operations could affect nearby communities. In October 2009, the Border Patrol reported that it was reevaluating its checkpoint performance measures, including quality of life measures. In December 2012, the DHS Center of Excellence completed a study for CBP on checkpoints. This study made several recommendations to Border Patrol on evaluating the impact of checkpoints on local communities using quantitative measures and with maintaining regular contact with the public to elicit opinions on



experiences with the checkpoint, both positive and negative. At the time, the Border Patrol noted it intended to develop quantitative measures on community impact, such as on public safety and quality of life, using information collected in the new checkpoint information system it was planning. Border Patrol also noted that it was considering the budgetary feasibility of (1) conducting a survey of checkpoint travelers to gather detailed information about the community and impact metrics that are of highest importance to the public and (2) implementing an expedited lane for regular and pre-approved travelers. In July 2014, the Border Patrol revised the expected completion date for its actions to address this recommendation to March 2015, noting that it planned to request ideas from the field commanders on what the agency could measure that would accurately depict the impact of checkpoints on the community. In June 2015, Border Patrol revised the expected completion date to September 2015. In September 2016, officials from Border Patrol’s Checkpoint Program Management Office said quality of life measures had not been implemented and they were not aware of any plans to develop and implement such measures. In its comments on this report (see app. IV), DHS stated that it expects to establish performance measures related to community impacts by February 28, 2018. As noted earlier in this report, residents and local law enforcement officials near checkpoints we spoke to for this review remain concerned about the effects checkpoints may have on their communities. Measuring performance, such as quality of life measures related to checkpoints, would give Border Patrol critical information on which to base decisions for improving checkpoint operations.

Recommendation 4: Use the information generated from the quality of life measures in conjunction with other relevant factors to inform resource allocations and address identified impacts.

Status: Open

In August 2009, we reported that while the Border Patrol’s national strategy cites the importance of assessing the community impact of Border Patrol operations, the implementation of such measures was lacking in terms of checkpoint operations. In October 2009, the Border Patrol reported that once it had completed an upgrade of its existing checkpoint data systems and had reevaluated its checkpoint performance measures, the agency would begin using information garnered by these performance measures to inform future resource allocation decisions. This was originally expected to be completed by September 30, 2010, but due to budgetary and other issues, the checkpoint system upgrades were



not yet completed as of June 2013. Border Patrol then reported to us in June 2013 that the redesigned and upgraded checkpoint information system was expected to be implemented in September 2014, but this system has not been developed or implemented, and in September 2016, officials from Border Patrol’s Checkpoint Program Management Office stated that they were not aware of any planned or completed actions to address this recommendation. In its comments on this report (see app. IV), DHS stated that it expects to establish performance measures related to community impacts by February 28, 2018, and that these measures will be used to inform resource allocation decisions. As noted earlier in this report, residents and local law enforcement officials near checkpoints we spoke to for this review remain concerned about the effects checkpoints may have on their communities. Measuring performance, such as quality of life measures related to checkpoints, would give Border Patrol critical information on which to base decisions for improving checkpoint operations.

Recommendation 5: Require that current and expected traffic volumes be considered by the Border Patrol when determining the number of inspection lanes at new permanent checkpoints, that traffic studies be conducted and documented, and that these requirements be explicitly documented in Border Patrol checkpoint design guidelines and standards.

Status: Closed – Implemented

In August 2009, we reported that Border Patrol did not conduct traffic studies when designing three recently constructed checkpoints. As a result, we could not determine if the Border Patrol complied with its checkpoint design guidelines to consider current and future traffic volumes when determining the number of inspection lanes at the three checkpoints. In the absence of documented traffic studies, the Border Patrol could not determine if the number of inspection lanes at each of these checkpoints was consistent with current and projected traffic volumes, or if a different number of lanes would have been more appropriate. On October 28, 2009, the Border Patrol finalized an addendum to the Border Patrol Facilities Design Standard, which requires the Border Patrol to acquire, document, and utilize traffic study data collected by the state Departments of Transportation regarding current and projected traffic volumes on roadways where permanent checkpoints are to be constructed. The traffic studies are to be documented by the Border Patrol and utilized as the baseline requirement to determine the number of inspection lanes at new permanent checkpoints, and therefore this recommendation has been closed as implemented.



Recommendation 6: In connection with planning for new or upgraded checkpoints, conduct a workforce planning needs assessment for checkpoint staffing allocations to determine the resources needed to address anticipated levels of illegal activity around the checkpoint.

Status: Open

In August 2009, we reported that Border Patrol’s checkpoint strategy to push illegal crossers and smugglers to areas around checkpoints—which could include nearby communities—underscores the need for the Border Patrol to ensure that it deploys sufficient resources and staff to these areas. In October 2009, Border Patrol reported that the agency was evaluating its checkpoint policy regarding the establishment of a new checkpoint or the upgrade of an old checkpoint, and checkpoint policy changes would be finalized by September 30, 2010. Border Patrol also reported that checkpoint system upgrades that capture data on checkpoint performance would help management determine future resource needs at checkpoints. In June 2013, Border Patrol reported that due to budget and other issues, the checkpoint system upgrade had not been completed, and the rewritten checkpoint data protocol had not been approved. In June 2013, Border Patrol reported that as part of the checkpoint study conducted by the DHS Center of Excellence, the Center created checkpoint simulation tools that would help inform resource allocations when determining the number of inspection lanes on current or new checkpoints. The Border Patrol agreed with the utility of such a model, but noted that the Border Patrol would need to purchase modeling software—a cost-prohibitive measure in the current budget environment. In the interim, Border Patrol is developing a formal workforce staffing model to identify staffing strategies for all Border Patrol duties. Border Patrol expected to implement this model for checkpoint staffing assignments in fiscal year 2014. However, in July 2014, Border Patrol reported that the Border Patrol’s Personnel Requirements Determination project was still being developed and that process would inform staffing at checkpoints, although the project is not specific to checkpoint staffing needs. As a result, Border Patrol revised its expected implementation date to September 2015. However, according to the Border Patrol official overseeing the project, subsequent changes in leadership and factors unrelated to checkpoints have affected the overall time frames for the Personnel Requirements Determination project. In September 2016, Border Patrol officials reported that the agency’s Personnel Requirements Determination process would not provide information on staffing needs until fiscal year 2017 or 2018. In its comments on this report (see app. IV), DHS stated that it expects to use information from the Personnel



Requirements Determination process to determine staffing requirements and address our recommendation by September 30, 2019. Given that local residents continue to express concerns about the impacts of checkpoints on communities, conducting a needs assessment when planning for a new or upgraded checkpoint could help better ensure that officials consider the potential impact of the checkpoint on the community and plan for a sufficient number of agents and resources.

Appendix II: Trends in Southwest Border Apprehensions, Fiscal Years 2012 through 2016


This appendix contains additional detail about trends in southwest border apprehensions from fiscal years 2012 through 2016, including trends in the:

• number of apprehensions by sector,

• distribution of apprehensions by sector and by distance from the border, and

• distribution of apprehensions by sector and by proximity to checkpoints.

From fiscal years 2012 through 2016, Border Patrol apprehended a total of almost 2 million individuals in southwest border sectors.1 The number of apprehensions over this period rose to a peak in fiscal year 2014, declined in fiscal year 2015, and rose again in fiscal year 2016. Over this 5-year period, about two-thirds of the apprehensions occurred in the Rio Grande Valley and Tucson sectors (42 percent and 23 percent, respectively), and the Rio Grande Valley sector accounted for an increasing percentage of total southwest border apprehensions over this time period (from 27 percent of all southwest border apprehensions in fiscal year 2012 to 46 percent of apprehensions in fiscal year 2016). As shown in figure 14, apprehensions also increased in five other sectors, but the other sectors represented consistently smaller percentages of all apprehensions over the 5-year period.2

1Border Patrol records data on each individual the agency arrests, including certain information about the individual (e.g., age, citizenship), the Global Positioning System (GPS) location of the arrest, and closest physical landmark (selected from among a dropdown list of preprogrammed locations). Upon arrest, Border Patrol determines whether foreign nationals are potentially removable on grounds of inadmissibility or deportability. Border Patrol also may encounter and, as appropriate, arrest U.S. citizens or foreign nationals with legal status in the United States for violating U.S. law, such as smuggling contraband. 2In addition to the Rio Grande Valley sector, apprehensions increased in the San Diego, Yuma, El Paso, Big Bend, and Del Rio sectors.


Apprehensions by Sector



Figure 14: Southwest Border Apprehensions by Border Patrol Sector, Fiscal Years 2012–2016

Note: These data exclude apprehensions with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

The Secretary of the Department of Homeland Security stated during testimony before the Senate Committee on Homeland Security and Governmental Affairs that apprehensions have dropped sharply since the beginning of 2017. He stated, for example, that Border Patrol apprehended approximately 1,000 unaccompanied alien children in March 2017 (a time of year he noted when apprehensions generally are



higher) compared to over 7,000 unaccompanied alien children in December 2016.3

As noted in this report, apprehensions overall for the southwest border increasingly occurred closer to the border. Table 6 shows the distribution for each sector of apprehensions by distance from the border during fiscal years 2012 through 2016.

Table 6: Southwest Border Apprehensions by Sector and by Distance from the Border for Fiscal Years (FY) 2012–2016

Sector Distance from the border

Percentage of sector’s apprehensions FY 2012 FY 2013 FY 2014 FY 2015 FY 2016

San Diego 0 mile to 0.5 mile 42 50 47 49 50 More than 0.5 mile to 1 mile 10 12 12 12 11 More than 1 mile to 5 miles 25 24 27 28 27 More than 5 miles to 10 miles 7 6 8 6 8 More than 10 miles to 20 miles 6 4 4 4 3 More than 20 miles 10 4 3 2 1

El Centro 0 mile to 0.5 mile 46 45 42 51 62 More than 0.5 mile to 1 mile 8 8 9 12 9 More than 1 mile to 5 miles 27 26 28 23 19 More than 5 miles to 10 miles 9 10 10 6 5 More than 10 miles to 20 miles 3 3 2 3 1 More than 20 miles 7 8 9 5 4

Yuma 0 mile to 0.5 mile 22 28 41 58 80 More than 0.5 mile to 1 mile 4 5 6 3 2 More than 1 mile to 5 miles 8 9 10 7 3 More than 5 miles to 10 miles 7 8 8 7 3 More than 10 miles to 20 miles 5 6 6 4 3 More than 20 miles 55 44 30 21 8

3Our analysis includes up to September 30, 2016, the last complete fiscal year available at the time of our review.

Distribution of Apprehensions by Sector and by Distance from the Border





Tucson 0 mile to 0.5 mile 14 15 26 26 21 More than 0.5 mile to 1 mile 5 5 7 7 7 More than 1 mile to 5 miles 25 28 25 23 26 More than 5 miles to 10 miles 17 15 10 10 11 More than 10 miles to 20 miles 14 14 12 12 12 More than 20 miles 25 23 20 21 23

El Paso 0 mile to 0.5 mile 39 29 34 33 44 More than 0.5 mile to 1 mile 8 11 11 11 13 More than 1 mile to 5 miles 16 19 20 21 18 More than 5 miles to 10 miles 8 7 5 6 6 More than 10 miles to 20 miles 10 11 8 8 5 More than 20 miles 19 24 21 21 14

Big Bend 0 mile to 0.5 mile 1 2 4 18 16 More than 0.5 miles to 1 mile 1 1 2 12 11 More than 1 mile to 5 miles 4 4 6 5 8 More than 5 miles to 10 miles 9 11 15 9 8 More than 10 miles to 20 miles 29 32 31 25 23 More than 20 miles 57 49 41 32 35

Del Rio 0 mile to 0.5 mile 12 22 41 30 32 More than 0.5 mile to 1 mile 5 6 6 5 7 More than 1 mile to 5 miles 19 19 16 18 20 More than 5 miles to 10 miles 15 12 8 11 10 More than 10 miles to 20 miles 21 19 12 16 14 More than 20 miles 29 23 17 20 17

Laredo 0 mile to 0.5 mile 26 30 37 32 30 More than 0.5 mile to 1 mile 6 7 7 8 9 More than 1 mile to 5 miles 14 13 12 16 17 More than 5 miles to 10 miles 8 8 7 7 6 More than 10 miles to 20 miles 9 7 5 5 6 More than 20 miles 38 36 33 33 32





Rio Grande Valley 0 mile to 0.5 mile 27 27 37 40 48 More than 0.5 mile to 1 mile 17 19 20 21 24 More than 1 mile to 5 miles 18 18 16 16 14 More than 5 miles to 10 miles 4 4 3 2 1 More than 10 miles to 20 miles 4 3 2 2 1 More than 20 miles 30 29 22 20 12


Notes: Percentages may not total to 100 because of rounding. These data exclude apprehensions with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

For fiscal years 2013 through 2016, the percent of apprehensions occurring at checkpoints varied by sector.4 We assigned each apprehension into one of four location categories based on whether the GPS coordinates for the event occurred close enough to the GPS coordinates for a checkpoint to be considered “at a checkpoint” and whether the event’s landmark corresponds to the nearest checkpoint landmark.5 Table 7 shows the distribution of apprehensions for each sector by location category during fiscal years 2013 through 2016, and the extent to which apprehensions were identified as checkpoint circumventions based on use of the “Circumvention App?” checkbox. Differences in sector apprehensions at checkpoints could depend in part on the number of checkpoints within a sector, the amount of time checkpoints are operational, and the extent to which sectors consistently apply guidance on how to enter data for apprehensions that are related to checkpoint operations.

4Our analysis of checkpoint apprehensions is limited to fiscal years 2013 through 2016 because Border Patrol began maintaining checkpoint lists for such historical analysis in fiscal year 2013. 5When recording data for an apprehension at a checkpoint, agents are to select the appropriate checkpoint location from a dropdown list of landmarks (landmark data field). Sector officials define landmarks and which landmarks are checkpoints within e3 for their respective sectors. Border Patrol began recording GPS coordinates for apprehensions and seizures in 2011, and prior to that, landmarks were used to denote the general area where an apprehension occurred.

Apprehensions at Checkpoints by Sector



Table 7: Southwest Border Checkpoint Apprehensions by Sector and by Location Category, Fiscal Years 2013–2016

Sector Location category

Number of apprehensions (percent of sector’s

apprehensions)a

Number of apprehensions with “Circumvention

Apprehension?” box checkedb

San Diego 1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

2,055 (1.79 percent)

4


9,623 (8.37 percent)

53

3. Occurred further than one-half mile from the checkpoint and is associated with the landmark for the nearest checkpoint

39 (0.03 percent)

0

4. Occurred further than one-half mile from the checkpoint and is associated with a landmark for a location other than the nearest checkpoint

103,216 (89.81 percent)

273

San Diego Total 114,933 330 El Centro


1,217 (2.03 percent)

8


341 (0.54 percent)

72


307 (0.49 percent)

86


60,773 (96.94 percent)

442

El Centro Total 62,692 608 Yuma


508 (1.53 percent)

0


153 (0.46 percent)

2


72 (0.22 percent)

0


32,408 (97.79 percent)

114

Yuma Total 33,141 116





apprehensions)a



Tucson 1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

1,890 (0.56 percent)

5


1,962 (0.59 percent)

28


187 (0.06 percent)

2


331,035 (98.79 percent)

629

Tucson Total 335,074 664 El Paso


2,234 (3.56 percent)

3


79 (0.13 percent)

1


56 (0.09 percent)

1


60,322 (96.22 percent)

130

El Paso Total 62,691 135 Big Bend


1,827 (9.56 percent)

0


351 (1.84 percent)

5


32 (0.17 percent)

1


16,906 (88.44 percent)

162

Big Bend Total 19,116 168





apprehensions)a



Del Rio 1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

436 (0.49 percent)

4


479 (0.54 percent)

1


91 (0.10 percent)

0


88,416 (98.87 percent)

60

Del Rio Total 89,422 65 Laredo


6,749 (4.07 percent)

1


1,708 (1.03 percent)

5


386 (0.23 percent)

1


156,828 (94.66 percent)

262

Laredo Total 165,671 269 Rio Grande Valley


14,669 (1.98 percent)

140


5,063 (0.68 percent)

425


576 (0.08 percent)

17


720,533 (97.26 percent)

25,610

Rio Grande Valley Total 740,841 26,192


Notes: Border Patrol began maintaining checkpoint lists used for our analysis in fiscal year 2013.



These data exclude apprehensions with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors. aPercentages may not total to 100 because of rounding. bThe number of apprehensions where the “Circumvention Apprehension?” box is checked is a subset within the overall number of apprehensions for each location category.

Appendix III: Trends in Southwest Border Seizures, Fiscal Years 2012 through 2016


This appendix contains additional detail about trends in southwest border seizures from fiscal years 2012 through 2016, including trends in the:

• number of seizures by type of contraband seized,

• number of seizures by sector,

• distribution of seizures by sector and by distance from the border,

• seizures related to Border Patrol checkpoints each available year by sector, and

• marijuana seizures at checkpoints by quantity seized.

Border Patrol seized almost 90,000 prohibited items in southwest border sectors from fiscal year 2012 through fiscal year 2016.1 Most of these seizures (92 percent) were narcotics, and 87 percent of narcotics seizures were marijuana. The remaining seizures were of firearms, ammunition, currency, or other property. As shown in table 8, the number of seizures on the southwest border generally decreased from fiscal year 2012 to fiscal year 2016, with the exceptions of slight rises in the amount of methamphetamines and heroin seized during this period.

Table 8: Type and Number of Southwest Border Seizures, Fiscal Years (FY) 2012–2016

Type of Property FY 2012 FY 2013 FY 2014 FY 2015 FY 2016 Total Narcotics

Marijuana 14,982 16,084 14,493 13,296 12,304 71,159 Cocaine 482 366 378 429 446 2,101 Hashish 274 287 321 268 258 1,408 Heroin 127 187 151 147 173 785 Methamphetamine 591 674 764 722 840 3,591 Other drugs 624 646 512 446 584 2,812

Firearms 578 533 486 319 343 2,259 Ammunition 629 632 547 364 393 2,565

1The number of seizures refers to a single property seizure, regardless of the size of the seizure. Different types of prohibited items seized from the same individual count as different seizures—for example, if marijuana, cocaine, and a firearm were seized from an individual, this would count as three seizures.


Seizures by Type of Contraband Seized



Type of Property FY 2012 FY 2013 FY 2014 FY 2015 FY 2016 Total Currency 608 632 424 448 430 2,542 Other Property 1 7 1 1 0 10 Total Seizures 18,896 20,048 18,077 16,440 15,771 89,232


Note: These data exclude seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

The greatest number of seizures during the 5 fiscal years occurred in the Tucson, Big Bend, and Rio Grande Valley sectors (34, 19, and 16 percent respectively). Collectively, these three sectors accounted for 69 percent of southwest border seizures from fiscal years 2012 through 2016. For all southwest border sectors except the Big Bend sector, the numbers of seizures decreased during this 5-year period. For example, the number of seizures in the Tucson sector decreased 12 percent, and the number of seizures in the Rio Grande Valley sector decreased 36 percent during this period. The number of seizures in the Big Bend sector increased 39 percent from fiscal years 2012 through 2016. Figure 15 shows the number of seizures from fiscal years 2012 through 2016 by sector.

Seizures by Sector



Figure 15: Southwest Border Seizures by Border Patrol Sector, Fiscal Years 2012–2016

Note: These data exclude seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors.

As noted in this report, the location where seizures occurred remained relatively stable from fiscal year 2012 through fiscal year 2016, with the majority of seizures occurring 10 miles or more from the southwest border. Table 9 shows the distribution of seizures for each sector by distance from the border during fiscal years 2012 through 2016.

Distribution of Seizures by Sector and by Distance from the Border



Table 9: Southwest Border Seizures by Sector and by Distance from the Border for Fiscal Years (FY) 2012–2016


Percentage of sector’s seizures FY 2012 FY 2013 FY 2014 FY 2015 FY 2016

San Diego 0 mile to 0.5 mile 1 2 2 2 3 More than 0.5 mile to 1 mile 0 1 1 1 1 More than 1 mile to 5 miles 2 2 3 5 10 More than 5 miles to 10 miles 10 8 11 15 20 More than 10 miles to 20 miles 50 54 53 45 34 More than 20 miles 36 33 31 31 32

El Centro 0 mile to 0.5 mile 19 17 11 15 28 More than 0.5 mile to 1 mile 23 16 2 4 4 More than 1 mile to 5 miles 5 6 5 5 4 More than 5 miles to 10 miles 5 6 4 3 3 More than 10 miles to 20 miles 6 3 4 6 2 More than 20 miles 41 53 73 68 59

Yuma 0 mile to 0.5 mile 8 9 2 2 2 More than 0.5 mile to 1 mile 0 0 0 0 0 More than 1 mile to 5 miles 2 2 2 3 2 More than 5 miles to 10 miles 3 4 4 4 5 More than 10 miles to 20 miles 14 21 54 36 37 More than 20 miles 73 64 39 54 54

Tucson 0 mile to 0.5 mile 10 8 9 10 9 More than 0.5 mile to 1 mile 4 4 4 3 3 More than 1 mile to 5 miles 21 19 20 18 17 More than 5 miles to 10 miles 17 16 13 13 14 More than 10 miles to 20 miles 13 15 15 15 14 More than 20 miles 35 37 39 42 43




Percentage of sector’s seizures FY 2012 FY 2013 FY 2014 FY 2015 FY 2016

El Paso 0 mile to 0.5 mile 5 2 7 7 8 More than 0.5 mile to 1 mile 1 2 0 3 4 More than 1 mile to 5 miles 5 6 7 10 9 More than 5 miles to 10 miles 9 4 4 6 4 More than 10 miles to 20 miles 5 5 4 6 7 More than 20 miles 75 81 77 69 69

Big Bend 0 mile to 0.5 mile 0 0 0 0 0 More than 0.5 mile to 1 mile 0 0 0 0 0 More than 1 mile to 5 miles 1 0 0 0 1 More than 5 miles to 10 miles 1 0 1 1 0 More than 10 miles to 20 miles 90 94 90 85 75 More than 20 miles 8 5 9 13 23

Del Rio 0 mile to 0.5 mile 19 19 11 10 4 More than 0.5 mile to 1 mile 6 6 7 2 1 More than 1 mile to 5 miles 13 16 10 11 8 More than 5 miles to 10 miles 19 27 35 42 42 More than 10 miles to 20 miles 8 8 11 8 10 More than 20 miles 36 23 26 28 34

Laredo 0 mile to 0.5 mile 17 15 18 20 13 More than 0.5 mile to 1 mile 5 5 7 6 6 More than 1 mile to 5 miles 14 14 9 8 8 More than 5 miles to 10 miles 4 2 3 2 1 More than 10 miles to 20 miles 2 1 2 2 2 More than 20 miles 58 63 61 61 71

Rio Grande Valley 0 mile to 0.5 mile 24 22 24 30 34 More than 0.5 mile to 1 mile 9 8 10 14 12 More than 1 mile to 5 miles 10 9 13 12 10 More than 5 miles to 10 miles 14 12 9 8 6 More than 10 miles to 20 miles 7 7 7 4 4 More than 20 miles 36 43 37 33 35




Notes: These data exclude seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors. Percentages may not total to 100 because of rounding.

For fiscal years 2013 through 2016, the percent of seizures occurring at checkpoints varied by sector. We assigned each seizure into one of four location categories based on whether the GPS coordinates for the event occurred close enough to the GPS coordinates for a checkpoint to be considered “at a checkpoint” and whether the event’s landmark corresponds to the nearest checkpoint landmark.2 Table 10 shows the distribution of seizures for each sector by checkpoint location category during fiscal years 2013 through 2016. Differences in sector seizures at checkpoints could depend in part on the number of checkpoints within a sector, the percent of time checkpoints are operational, and the extent to which sectors consistently apply guidance on how to enter data for seizures that are related to checkpoint operations.

Table 10: Southwest Border Seizures by Sector and by Location Category, Fiscal Years 2013–2016


Number of seizures (percent of sector’s seizures)a

San Diego 1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

3,400 (69.8 percent)


321 (6.6 percent)


26 (0.5 percent)


1,127 (23.1 percent)

San Diego Total 4,874

2When recording data for a seizure at a checkpoint, agents are to select the appropriate checkpoint location from a dropdown list of landmarks (landmark data field). Sector officials define landmarks and which landmarks are checkpoints within e3 for their respective sectors. Border Patrol began recording GPS coordinates for apprehensions and seizures in 2011, and prior to that, landmarks were used to denote the general area where a seizure occurred.

Seizures at Checkpoints by Sector





El Centro 1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

742 (55.5 percent)


7 (0.5 percent)


13 (1.0 percent)


575 (43.0 percent)

El Centro Total 1,337 Yuma


3,501 (57.4 percent)


141 (2.3 percent)


140 (2,3 percent)


2,322 (38.0 percent)

Yuma Total 6,104 Tucson


2,206 (9.0 percent)


219 (0.9 percent)


82 (0.3 percent)


21,907 (89.7 percent)

Tucson Total 24,414





El Paso 1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

1,817 (43.3 percent)


29 (0.7 percent)


63 (1.5 percent)


2,287 (54.5 percent)

El Paso Total 4,196 Big Bend


13,354 (95.2 percent)


49 (0.3 percent)


13 (0.1 percent)


613 (4.4 percent)

Big Bend Total 14,029 Del Rio


729 (49.4 percent)


47 (3.2 percent)


30 (2.0 percent)


669 (45.4 percent)

Del Rio Total 1,475





Laredo 1. Occurred one-half mile or less from the checkpoint and is associated with the landmark for the nearest checkpoint

1,252 (42.7 percent)


139 (4.7 percent)


154 (5.3 percent)


1,385 (47.3 percent)

Laredo Total 2,930 Rio Grande Valley


3,448 (31.4 percent)


230 (2.1 percent)


61 (0.6 percent)


7,238 (66.0 percent)

Rio Grande Valley Total 10,997


Note: Border Patrol began maintaining checkpoint lists used for our analysis in fiscal year 2013. These data exclude seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors. aPercentages may not total to 100 because of rounding.

Most southwest border seizures were narcotics, and most narcotics seizures were marijuana. As noted in this report, marijuana seizures at checkpoints were often for smaller quantities compared to marijuana seizures at non-checkpoint locations. Table 11 shows that about 67 percent of marijuana seizures at checkpoints were for quantities less than or equal to 1 ounce, whereas the quantities seized at non-checkpoint locations were often larger. For example, more than three-quarters of marijuana seizures at non-checkpoint locations were of over 50 pounds (25,792 out of 33,477 seizures).

Marijuana Seizures by Quantity Seized



Table 11: Southwest Border Marijuana Seizures by Quantity Seized at Checkpoints and at Non-Checkpoint Locations, Fiscal Years 2013–2016

Quantity of marijuana seized Number of marijuana seizures per location categorya At checkpoints

(occurred one-half mile or less from the checkpoint and is associated with

the landmark for the nearest checkpoint)

At non-checkpoint locations (occurred further than one-half mile

from the checkpoint and is associated with a landmark for a

location other than the nearest checkpoint)

Unspecifiedb 540 101 1 ounce or less 14,299 1,789 More than 1 ounce, less than 1 pound 3,348 1,012 1 pound or more, less than 50 pounds 1,723 4,783 50 pounds or more, less than 250 pounds 1,098 15,911 250 pounds or more 366 9,881 Total 21,374 33,477


Note: These data exclude seizures with missing GPS coordinates or with GPS coordinates corresponding to a location outside the United States or outside the boundaries of the southwest border sectors. aThis table focuses on seizures for which the data on locations (landmark and GPS coordinates) consistently indicated the location was either at a checkpoint or at a non-checkpoint location. As discussed in this report, the relationship between some seizures and checkpoints is unclear because such seizures either (a) occurred one-half mile or less from the checkpoint and were associated with a location other than the nearest checkpoint or (b) occurred farther than one-half mile from the checkpoint but were associated with the checkpoint location. Such seizures represent a small percentage (2.5 percent) of total seizures over the time period we analyzed. bThe category for unspecified subject refers to a seizure for which a Border Patrol agent did not attribute the seizure to a particular individual who was arrested or apprehended in conjunction with the seizure. Some examples when this might occur are if there were multiple people in the vehicle that contained the contraband or if Border Patrol found the contraband while it was unattended.

Appendix IV: Comments from the Department of Homeland Security



Appendix V: GAO Contact and Staff Acknowledgments


Rebecca Gambler, (202) 512-8777 or [email protected]

In addition to the contact named above, Adam Hoffman (Assistant Director), David Alexander, Alana Finley, Eric Hauswirth, Monica Kelly, John Mingus, Sasan J. “Jon” Najmi, Christine San, Adam Vogt, and Tomas Wind made significant contributions to this report.

Appendix V: GAO Contact and Staff Acknowledgments

GAO Contact

Staff Acknowledgments

(100988)


The Government Accountability Office, the audit, evaluation, and investigative arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability.

The fastest and easiest way to obtain copies of GAO documents at no cost is through GAO’s website (http://www.gao.gov). Each weekday afternoon, GAO posts on its website newly released reports, testimony, and correspondence. To have GAO e-mail you a list of newly posted products, go to http://www.gao.gov and select “E-mail Updates.”

The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, http://www.gao.gov/ordering.htm.

Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537.

Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information.

Connect with GAO on Facebook, Flickr, LinkedIn, Twitter, and YouTube. Subscribe to our RSS Feeds or E-mail Updates. Listen to our Podcasts. Visit GAO on the web at www.gao.gov and read The Watchblog.

Contact:

Website: http://www.gao.gov/fraudnet/fraudnet.htm E-mail: [email protected] Automated answering system: (800) 424-5454 or (202) 512-7470

Orice Williams Brown, Managing Director, [email protected], (202) 512-4400, U.S. Government Accountability Office, 441 G Street NW, Room 7125, Washington, DC 20548

Chuck Young, Managing Director, [email protected], (202) 512-4800 U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, DC 20548

James-Christian Blockwood, Managing Director, [email protected], (202) 512-4707 U.S. Government Accountability Office, 441 G Street NW, Room 7814, Washington, DC 20548

GAO’s Mission

Obtaining Copies of GAO Reports and Testimony Order by Phone

Connect with GAO

To Report Fraud, Waste, and Abuse in Federal Programs

Congressional Relations

Public Affairs

Strategic Planning and External Liaison

Please Print on Recycled Paper.

http://www.gao.gov/

http://www.gao.gov/

http://www.gao.gov/ordering.htm

http://facebook.com/usgao

http://flickr.com/usgao

http://www.linkedin.com/company/us-government?trk=cp_followed_name_us-government

http://twitter.com/usgao

http://youtube.com/usgao

http://www.gao.gov/feeds.html

http://www.gao.gov/subscribe/index.php

http://www.gao.gov/podcast/watchdog.html

http://www.gao.gov/

http://blog.gao.gov/

http://www.gao.gov/fraudnet/fraudnet.htm





Download - GAO-18-50, BORDER PATROL: Issues Related to Agent ...Issues Related to Agent Deployment Strategy and Immigration Checkpoints . What GAO Found . According to U.S. Border Patrol (Border

Top Related